A data processing method, device, apparatus, and computer-readable storage medium

Abstract

The application discloses a data processing method, device and equipment and a computer readable storage medium. The method comprises the following steps: determining a perception coverage area and a perception blank area with perception data in a simulation backbone road; at a simulation starting moment, generating a first virtual simulation vehicle in the perception blank area according to the area relationship between the perception coverage area and the perception blank area; at a simulation reproduction stage, outputting a reproduced simulation driving behavior corresponding to the perception data in the perception coverage area, and outputting a virtual simulation driving behavior of a second virtual simulation vehicle in the perception blank area according to the first virtual simulation vehicle; and at a simulation prediction stage, outputting a predicted simulation driving behavior of a third virtual simulation vehicle on the simulation backbone road according to the virtual simulation driving behavior and the reproduced simulation driving behavior. The application can improve the reproduction accuracy and prediction accuracy of the simulation backbone road. The embodiment of the application can be applied to the fields of maps and automatic driving.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to a data processing method, apparatus, device, and computer-readable storage medium. Background Technology

[0002] As society develops, traffic on main roads becomes increasingly complex. Therefore, relevant personnel not only need to understand the real-time traffic conditions on main roads, but also need to predict the future traffic conditions.

[0003] Since not all sections of a main road have sensing data, existing technologies can only simulate the portion of the main road with sensing data. Therefore, existing technologies cannot accurately reproduce the entire condition of the main road, thus reducing the accuracy of the simulated main road. Furthermore, this low accuracy leads to low prediction accuracy in the simulated main road. Summary of the Invention

[0004] This application provides a data processing method, apparatus, device, and computer-readable storage medium, which can improve the accuracy of reproducing simulated main roads and thus improve the accuracy of predicting simulated main roads.

[0005] One embodiment of this application provides a data processing method, including:

[0006] In the driving simulation system, the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area are determined in the simulated main road.

[0007] At the start of the simulation, based on the regional relationship between the sensing coverage area and the sensing blank area, the first virtual simulation vehicle is generated in the sensing blank area.

[0008] During the simulation reproduction phase, which is later than the start of the simulation, the reproduced simulated driving behavior corresponding to the perception data is output in the perception coverage area. Based on the first virtual simulated vehicle, the virtual simulated driving behavior of the second virtual simulated vehicle is output in the perception blank area. The second virtual simulated vehicle includes the first virtual simulated vehicle.

[0009] In the simulation prediction phase, which is later than the simulation reproduction phase, the predicted simulation driving behavior of the third virtual simulation vehicle is output on the simulation main road based on the virtual simulation driving behavior and the reproduced simulation driving behavior.

[0010] One embodiment of this application provides a data processing apparatus, including:

[0011] The area determination module is used in the driving simulation system to determine the perception coverage area with perception data and the perception blank area that does not overlap with the perception coverage area in the simulated main road.

[0012] The vehicle generation module is used to generate the first virtual simulation vehicle in the perception blank area at the start of the simulation, based on the regional relationship between the perception coverage area and the perception blank area.

[0013] The first output module is used to output the reproduced simulated driving behavior corresponding to the perception data in the perception coverage area during the simulation reproduction phase, which is later than the simulation start time; and to output the virtual simulated driving behavior of the second virtual simulated vehicle in the perception blank area based on the first virtual simulated vehicle; the second virtual simulated vehicle includes the first virtual simulated vehicle.

[0014] The second output module is used to output the predicted simulated driving behavior of the third virtual simulated vehicle on the simulated main road, based on the virtual simulated driving behavior and the reproduced simulated driving behavior, in the simulation prediction stage, which is later than the simulation reproduction stage.

[0015] The vehicle generation module includes:

[0016] The first determining unit is used to determine the initial traffic state corresponding to the sensing blank area based on the regional relationship between the sensing coverage area and the sensing blank area.

[0017] The first generation unit is used to generate the first virtual simulation vehicle in the perception blank area based on the initial traffic conditions;

[0018] Among them, the vehicle generation module is also used to obtain the initial perception data of the perception coverage area at the start of the simulation from the perception data;

[0019] The vehicle generation module is also used to generate an initial reproduction simulation vehicle in the perception coverage area based on the initial perception data; the initial reproduction simulation vehicle is the reproduction simulation vehicle corresponding to the reproduction simulation driving behavior.

[0020] The total number of sensing coverage areas is P, where P is a positive integer;

[0021] The first defined unit includes:

[0022] The first acquisition subunit is used to acquire, within the P sensing coverage areas, the sensing coverage area that is connected to the sensing blank area and located downstream of the sensing blank area, as the first sensing coverage area, if the regional relationship between the P sensing coverage areas and the sensing blank area is an upstream regional relationship or a midstream regional relationship; the upstream regional relationship is used to characterize that the sensing blank area is an upstream region of the P sensing coverage areas; the midstream regional relationship is used to characterize that the sensing blank area is located between two sensing coverage areas in the P sensing coverage areas.

[0023] The second acquisition subunit is used to acquire the first initial sensing data corresponding to the first sensing coverage area from the initial sensing data;

[0024] The first determining subunit is used to determine the initial traffic state corresponding to the sensing blank area based on the first initial sensing data if the first initial sensing data meets the state setting conditions for the sensing blank area.

[0025] The second determining subunit is used to obtain the first historical data of the sensing blank area if the first initial sensing data does not meet the state setting conditions, and determine the initial traffic state based on the first historical data.

[0026] The first sensing coverage area includes A simulated lanes; A is a positive integer; the A simulated lanes include simulated lane E. f f is a positive integer and f is less than or equal to A;

[0027] The first determining unit also includes:

[0028] The third acquisition subunit is used to acquire the location of lane E in the simulation lane from the first initial perception data. f The simulated vehicle G in the simulation f Reproduce simulated vehicle G f This belongs to the initial reproduction simulation vehicle;

[0029] The third determining subunit is used to reproduce the simulated vehicle G. f If the total number is at least two, then the first initial sensing data is determined to meet the state setting conditions;

[0030] The third determining subunit is also used to reproduce the simulated vehicle G. f If the total number is less than or equal to 1, then the simulation lane E is determined. f If A simulated lanes are considered as empty lanes, then the first initial perception data does not meet the state setting conditions.

[0031] The first determining subunit is specifically used to determine at least two reproducible simulation vehicles G. f The distance between any two adjacent simulated vehicles in the simulation.

[0032] The first determining subunit is further specifically used to determine at least two replica simulation vehicles G based on the vehicle spacing between every two adjacent replica simulation vehicles. f The average vehicle spacing;

[0033] The first defined subunit is also specifically used to integrate at least two simulated vehicles G. f The average vehicle spacing is determined as the simulated lane E. f The corresponding average vehicle spacing;

[0034] The first determining subunit is also specifically used to determine the vehicle density corresponding to the perception blank area based on the average vehicle spacing corresponding to A simulated lanes.

[0035] The first determining subunit is also specifically used to determine the initial traffic state corresponding to the sensing blank area in the first basic traffic map corresponding to the first sensing coverage area, based on the vehicle density.

[0036] The second determining subunit is specifically used to obtain the first starting historical data corresponding to the simulation start time from the first historical data if the first historical data is not an empty set, and to determine the first starting historical data as the starting traffic state.

[0037] The second determining subunit is also specifically used to determine the initial traffic state based on the target traffic state in the second basic traffic map corresponding to the perceived blank area if the first historical data is an empty set.

[0038] The total number of sensing coverage areas is M, where M is a positive integer;

[0039] The first defined unit includes:

[0040] The fourth acquisition subunit is used to acquire the second historical data of the sensing blank area if the regional relationship between the M sensing coverage areas and the sensing blank area is a downstream regional relationship; the downstream regional relationship is used to characterize that the sensing blank area is a downstream region of the M sensing coverage areas.

[0041] The fifth acquisition subunit is used to acquire the second starting historical data corresponding to the simulation start time from the second historical data if the second historical data is not an empty set, and to determine the second starting historical data as the starting traffic state corresponding to the perception blank area.

[0042] The sixth acquisition subunit is used to acquire, in the M sensing coverage areas, the sensing coverage area that is connected to the sensing blank area and located upstream of the sensing blank area, as the upstream sensing coverage area, if the second historical data is an empty set, and to determine the starting traffic state based on the upstream sensing coverage area.

[0043] The sixth acquisition subunit is specifically used to acquire the second initial sensing data corresponding to the upstream sensing coverage area from the initial sensing data.

[0044] The sixth acquisition subunit is also specifically used to determine the initial traffic state based on the second initial perception data if the second initial perception data meets the state setting conditions for the perception blank area.

[0045] The sixth acquisition subunit is also specifically used to determine the initial traffic state based on the target traffic state in the third basic traffic map corresponding to the perceived blank area if the second initial sensing data does not meet the state setting conditions.

[0046] The first generation unit includes:

[0047] The fourth determination subunit is used to determine the average vehicle spacing corresponding to the perception blank area based on the initial traffic conditions.

[0048] The first generation subunit is used to generate a first virtual simulation vehicle in the perception blank area if the regional relationship is an upstream regional relationship or a midstream regional relationship, based on the average vehicle spacing corresponding to the perception blank area, the downstream edge of the perception blank area, and the direction opposite to the driving direction of the simulated main road. The upstream regional relationship is used to characterize that the perception blank area is an upstream area of ​​the perception coverage area. The midstream regional relationship is used to characterize that the total number of perception coverage areas is at least two, and the perception blank area is located between two perception coverage areas in at least two perception coverage areas.

[0049] The second generation subunit is used to generate a first virtual simulation vehicle in the perception blank area based on the average vehicle spacing corresponding to the perception blank area, the upstream edge of the perception blank area, and the driving direction if the regional relationship is a downstream regional relationship; the downstream regional relationship is used to characterize that the perception blank area is a downstream region of the perception coverage area.

[0050] Among them, the regional relationship is the upstream regional relationship; the upstream regional relationship is used to characterize that the perceived blank area is the upstream region of the perceived coverage area;

[0051] The first output module includes:

[0052] The second determining unit is used to determine the first vehicle generation sub-region in the perception blank region based on the upstream edge of the perception blank region and the vehicle generation line in the perception blank region; the vehicle generation line is perpendicular to the driving direction of the simulated main road.

[0053] The first acquisition unit is used to acquire downstream sensing data corresponding to the second sensing coverage area from the sensing data; the second sensing coverage area refers to the area in the sensing coverage area that is connected to the sensing blank area and located downstream of the sensing blank area.

[0054] The second generation unit is used to generate a fourth virtual simulation vehicle in the first vehicle generation sub-region based on downstream sensing data, and to determine the first virtual simulation vehicle and the fourth virtual simulation vehicle as the second virtual simulation vehicle.

[0055] The third determining unit is used to determine a first vehicle removal line perpendicular to the driving direction of the simulated main road in the perception blank area; the first vehicle removal line is located in the downstream area of ​​the first vehicle generation sub-area, and the distance between the first vehicle removal line and the downstream edge of the perception blank area is less than the distance between the first starting vehicle and the downstream edge of the perception blank area; the first starting vehicle refers to the virtual simulation vehicle that is closest to the downstream edge of the perception blank area among the first virtual simulation vehicles.

[0056] The first output unit is used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area according to the autonomous driving model corresponding to the perception blank area and the first vehicle removal line; the first vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle in the second virtual simulation vehicle that drives to the first vehicle removal line.

[0057] The second generation unit includes:

[0058] The third generation sub-unit is used to generate a fourth virtual simulation vehicle in the first vehicle generation sub-region based on the downstream sensing data if the downstream sensing data meets the state setting conditions for the sensing blank area.

[0059] The fourth generation sub-unit is used to obtain the third historical data of the perception blank area if the downstream perception data does not meet the state setting conditions, and generate the fourth virtual simulation vehicle in the first vehicle generation sub-region based on the third historical data.

[0060] The second sensing coverage area includes Z simulated lanes; Z is a positive integer; the Z simulated lanes include simulated lane X y y is a positive integer and y is less than or equal to Z;

[0061] The second generation unit also includes:

[0062] The seventh acquisition subunit is used to acquire the location of lane X in the simulation lane from the downstream sensing data. y The simulated vehicle U in the middle y Reproducing the simulated vehicle U y These are the simulated vehicles that correspond to the simulated driving behavior.

[0063] The fifth defined subunit is used to reproduce the simulated vehicle U. y If the total number is at least two, then the downstream sensing data is determined to meet the state setting conditions;

[0064] The fifth determining subunit is also used to reproduce the simulated vehicle U. y If the total number is less than or equal to 1, then the simulation lane X is determined. yIf Z simulated lanes are considered as empty lanes, then the downstream sensing data does not meet the state setting conditions.

[0065] The third generation subunit is specifically used to determine at least two replica simulation vehicles U. y Based on the corresponding reproduction timestamps, the simulation lane X is determined using at least two reproduction timestamps. y The corresponding average reproduction time interval is determined based on the average reproduction time intervals corresponding to the Z simulated lanes, thus determining the average reproduction time interval corresponding to the first vehicle generation sub-region.

[0066] The third generation sub-unit is also specifically used to determine at least two replica simulation vehicles U. y Based on the corresponding vehicle speeds, determine the simulated lane X using at least two vehicle speeds. y The average vehicle speed is determined based on the average vehicle speeds of the Z simulated lanes, which determines the average vehicle speed of the first vehicle generation sub-region.

[0067] The third generation sub-unit is also specifically used to generate a fourth virtual simulation vehicle in the first vehicle generation sub-region with an initial speed equal to the average vehicle speed corresponding to the first vehicle generation sub-region, based on the average reproduction time interval corresponding to the first vehicle generation sub-region.

[0068] The fourth generation sub-unit is specifically used to obtain reproduced historical data from the third historical data if the third historical data is not an empty set, and to generate a fourth virtual simulation vehicle in the first vehicle generation sub-region based on the reproduced historical data.

[0069] The fourth generation sub-unit is specifically used to generate a fourth virtual vehicle in the first vehicle generation sub-region if the third historical data is an empty set, based on the target traffic state in the fourth basic traffic map corresponding to the perceived blank area.

[0070] The first output module further includes:

[0071] The second acquisition unit is used to acquire the initial autonomous driving model corresponding to the perception blank area and to acquire the third historical data of the perception blank area.

[0072] The parameter adjustment unit is used to adjust the parameters in the initial autonomous driving model corresponding to the perception blank area based on the third historical data if the third historical data is not an empty set, so as to obtain the autonomous driving model corresponding to the perception blank area.

[0073] The parameter adjustment unit is also used to adjust the parameters in the initial autonomous driving model corresponding to the perceived blank area according to the road type corresponding to the perceived blank area if the third historical data is an empty set, so as to obtain the autonomous driving model corresponding to the perceived blank area.

[0074] The first output module further includes:

[0075] The fourth determining unit is used to determine the virtual simulation vehicle in the second virtual simulation vehicle that is closest to the downstream edge of the perception blank area as the downstream vehicle of the perception blank area;

[0076] The fifth determining unit is used to determine the maximum vehicle speed of downstream vehicles based on downstream sensing data.

[0077] The sixth determining unit is used to determine the virtual simulation vehicles in the second virtual simulation vehicle other than the downstream vehicles as the first upstream vehicles;

[0078] The seventh determining unit is used to determine the maximum vehicle speed of the first upstream vehicle based on the road type corresponding to the perceived blank area.

[0079] The first output unit is specifically used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the autonomous driving model corresponding to the perception blank area, the first vehicle removal line, the maximum vehicle speed of the first upstream vehicle, and the maximum vehicle speed of the downstream vehicle.

[0080] The fifth determining unit includes:

[0081] The sixth determining subunit is used to determine the second upstream vehicle as the replica simulation vehicle that is closest to the upstream edge of the second sensing coverage area among the replica simulation vehicles in the second sensing coverage area if the downstream sensing data indicates that there is a replica simulation vehicle in the second sensing coverage area.

[0082] The seventh determining subunit is used to determine the vehicle speed of the second upstream vehicle as the maximum vehicle speed of the downstream vehicle; the simulation lane to which the second upstream vehicle belongs is the same as the simulation lane to which the downstream vehicle belongs.

[0083] The eighth acquisition subunit is used to acquire the historical vehicle speed from the third historical data if the downstream sensing data indicates that there is no reproduced simulation vehicle in the second sensing coverage area and the third historical data of the sensing blank area is not an empty set, and to determine the historical vehicle speed as the maximum vehicle speed of the downstream vehicle.

[0084] The eighth determining subunit is used to determine the maximum vehicle speed of the downstream vehicle based on the road type corresponding to the sensing blank area if the downstream sensing data indicates that there is no reproduced simulated vehicle within the second sensing coverage area and the third historical data of the sensing blank area is an empty set.

[0085] Among them, the regional relationship is the midstream regional relationship; the midstream regional relationship is used to represent that the total number of sensing coverage areas is at least two, and the sensing blank area is located between two sensing coverage areas in the at least two sensing coverage areas;

[0086] The first output module includes:

[0087] The eighth determining unit is used to determine the replica simulation vehicle that drives to the perception blank area among the replica simulation vehicles in the third perception coverage area as the fifth virtual simulation vehicle; the replica simulation vehicle in the third perception coverage area is generated by perception data; the third perception coverage area refers to the area that is connected to the perception blank area and located downstream of the perception blank area among at least two perception coverage areas.

[0088] The eighth determining unit is also used to determine the first virtual simulation vehicle and the fifth virtual simulation vehicle as the second virtual simulation vehicle;

[0089] The ninth determining unit is used to determine a second vehicle removal line perpendicular to the driving direction of the simulated main road in the sensing blank area; the distance between the second vehicle removal line and the downstream edge of the sensing blank area is less than the distance between the second starting vehicle and the downstream edge of the sensing blank area; the second starting vehicle refers to the virtual simulated vehicle in the first virtual simulated vehicle that is closest to the downstream edge of the sensing blank area.

[0090] The second output unit is used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area according to the autonomous driving model corresponding to the perception blank area and the second vehicle removal line; the second vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle that drives to the second vehicle removal line.

[0091] Among them, the regional relationship is the downstream regional relationship; the downstream regional relationship is used to characterize the sensing blank area as the downstream region of the sensing coverage area;

[0092] The first output module includes:

[0093] The tenth determining unit is used to determine the replica simulation vehicle that drives to the perception blank area among the replica simulation vehicles in the upstream perception coverage area as the sixth virtual simulation vehicle; the replica simulation vehicle in the upstream perception coverage area is generated by perception data; the upstream perception coverage area refers to the area in the perception coverage area that is connected to the perception blank area and located upstream of the perception blank area.

[0094] The tenth determining unit is also used to determine the first virtual simulation vehicle and the sixth virtual simulation vehicle as the second virtual simulation vehicle;

[0095] The third output unit is used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area according to the autonomous driving model corresponding to the perception blank area and the downstream edge of the perception blank area; the downstream edge of the perception blank area is used to instruct the driving simulation system to delete the virtual simulation vehicle in the second virtual simulation vehicle that drives to the downstream edge of the perception blank area.

[0096] The second output module includes:

[0097] The vehicle determination unit is used to determine the virtual simulation vehicle corresponding to the simulation driving behavior that meets the simulation prediction conditions in the virtual simulation driving behavior as the seventh virtual simulation vehicle; the seventh virtual simulation vehicle belongs to the second virtual simulation vehicle.

[0098] The target determination unit is used to identify the simulated driving behavior that meets the simulation prediction conditions in the simulated driving behavior as the target simulated vehicle; the target simulated vehicle is generated based on perception data.

[0099] The third generation unit is used to determine the second vehicle generation sub-region in the simulated main road based on the upstream edge of the simulated main road.

[0100] The fourth generation unit is used to generate an eighth virtual simulation vehicle in the second vehicle generation sub-region based on the fourth historical data of the perception region to which the second vehicle generation sub-region belongs; the perception region to which the second vehicle generation sub-region belongs belongs to the perception coverage area or the perception blank area.

[0101] The vehicle determination unit is also used to determine the seventh virtual simulation vehicle, the target reproduction simulation vehicle, and the eighth virtual simulation vehicle as the third virtual simulation vehicle;

[0102] The fourth output unit is used to output the predicted simulated driving behavior of the third virtual simulated vehicle on the simulated main road based on the autonomous driving model corresponding to the simulated main road and the downstream edge of the simulated main road; the downstream edge of the simulated main road is used to instruct the driving simulation system to delete the virtual simulated vehicle that drives to the downstream edge of the simulated main road in the third virtual simulated vehicle.

[0103] The third generation unit includes:

[0104] The ninth determining sub-unit is used to determine the first vehicle generation sub-region corresponding to the perception blank area as the second vehicle generation sub-region of the simulation main road if the upstream edge of the simulated main road belongs to the perception blank area; the first vehicle generation sub-region is used to generate virtual simulation vehicles belonging to the second virtual simulation vehicles in the simulation reproduction stage.

[0105] The tenth determining sub-unit is used to determine a second vehicle generation sub-region in the simulated main road if the upstream edge of the simulated main road belongs to the sensing coverage area; the vehicle generation line in the simulated main road is perpendicular to the driving direction of the simulated main road.

[0106] The fourth generation unit includes:

[0107] The fifth generation sub-unit is used to obtain predicted historical data from the fourth historical data if the fourth historical data of the perception area to which the second vehicle generation sub-region belongs is not an empty set, and to generate the eighth virtual simulation vehicle in the second vehicle generation sub-region based on the predicted historical data.

[0108] The sixth generation sub-unit is used to generate an eighth virtual vehicle in the second vehicle generation sub-region if the fourth historical data is an empty set and the perception region to which the second vehicle generation sub-region belongs is a perception blank region.

[0109] The seventh generation sub-unit is used to generate an eighth virtual vehicle in the second vehicle generation sub-region based on the perception data if the fourth historical data is an empty set and the perception region to which the second vehicle generation sub-region belongs is a perception coverage region.

[0110] The second output module also includes:

[0111] The downstream determination unit is used to determine the virtual simulation vehicle that is closest to the downstream edge of the simulated main road among the third virtual simulation vehicles as the second downstream vehicle of the simulated main road.

[0112] The speed determination unit is used to determine the maximum vehicle speed of the second downstream vehicle based on the fifth historical data of the sensing area to which the second downstream vehicle belongs; the sensing area to which the second downstream vehicle belongs belongs to either a sensing blank area or a sensing coverage area.

[0113] The downstream determination unit is also used to determine the virtual simulation vehicles other than the second downstream vehicles in the third virtual simulation vehicle as the third upstream vehicles;

[0114] The speed determination unit is also used to determine the maximum vehicle speed of the third upstream vehicle based on the road type corresponding to the perception area to which the third upstream vehicle belongs.

[0115] The fourth output unit is specifically used to output the predicted simulated driving behavior of the third virtual simulated vehicle on the simulated main road based on the autonomous driving model corresponding to the simulated main road, the downstream edge of the simulated main road, the maximum vehicle speed of the third upstream vehicle, and the maximum speed of the second downstream vehicle.

[0116] This application provides a computer device, including: a processor, a memory, and a network interface;

[0117] The processor is connected to the memory and the network interface, wherein the network interface is used to provide data communication functions, the memory is used to store computer programs, and the processor is used to call the computer programs so that the computer device executes the methods in the embodiments of this application.

[0118] One aspect of this application provides a computer-readable storage medium storing a computer program adapted for loading by a processor and executing the methods described in this application.

[0119] One aspect of this application provides a computer program product, which includes a computer program stored in a computer-readable storage medium; a processor of a computer device reads the computer program from the computer-readable storage medium and executes the computer program, causing the computer device to perform the method described in this application.

[0120] In this embodiment, the computer device can generate a first virtual simulated vehicle in the perceptual blank area at the simulation start time, based on the regional relationship between the perceptual coverage area and the perceptual blank area. The perceptual coverage area includes regions with perceptual data on the simulated main road, and the perceptual blank area includes regions on the simulated main road that do not overlap with the perceptual coverage area. Obviously, the perceptual blank area does not include perceptual data. Further, in the simulation reproduction stage, which is later than the simulation start time, the reproduced simulated driving behavior corresponding to the perceptual data is output in the perceptual coverage area, and the virtual simulated driving behavior of a second virtual simulated vehicle is output in the perceptual blank area based on the first virtual simulated vehicle. Clearly, in the simulation reproduction stage, the perceptual coverage area outputs perceptual data, and the perceptual blank area outputs virtual data. Further, in the simulation prediction stage, which is later than the simulation reproduction stage, the predicted simulated driving behavior of a third virtual simulated vehicle is output on the simulated main road based on the virtual simulated driving behavior and the reproduced simulated driving behavior. Clearly, in the simulation prediction stage, not only the perceptual blank area outputs prediction data, but the perceptual coverage area also outputs prediction data. As can be seen from the above, in the embodiments of this application, the perception blank areas without perception data are described by simulated vehicles at the simulation start time and in the simulation reproduction stage, which can improve the accuracy of the driving simulation system in reproducing the simulated main road. In addition, the embodiments of this application describe the simulated main road in different simulation stages separately, which can improve the prediction accuracy of the driving simulation system for the simulated main road. Attached Figure Description

[0121] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0122] Figure 1 This is a schematic diagram of a system architecture provided in an embodiment of this application;

[0123] Figure 2 This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 1 ;

[0124] Figure 3 This is a schematic diagram of a sensing coverage area and a sensing blank area provided in an embodiment of this application;

[0125] Figure 4a This is a schematic diagram of a scenario in which a first virtual simulation vehicle is generated in a downstream sensing area, as provided in an embodiment of this application.

[0126] Figure 4b This is a flowchart illustrating a method for generating virtual simulated vehicles in the downstream sensing region at the start of simulation, as provided in an embodiment of this application.

[0127] Figure 4c This is a schematic diagram of a scenario in which virtual simulated driving behavior is output in the downstream area of ​​perception, provided by an embodiment of this application;

[0128] Figure 4d This is a flowchart illustrating a method for generating virtual simulated driving behavior in the downstream perception area during the simulation reproduction stage, as provided in an embodiment of this application.

[0129] Figure 5 This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 2 ;

[0130] Figure 6a This is a schematic diagram of a scenario in which a first virtual simulation vehicle is generated in the upstream sensing area, as provided in an embodiment of this application.

[0131] Figure 6b This is a flowchart illustrating a method for generating virtual simulated vehicles in the upstream / midstream perception region at the start of simulation, as provided in an embodiment of this application.

[0132] Figure 6c This is a schematic diagram of a scenario in which virtual simulated driving behavior is output in the upstream region of perception, provided by an embodiment of this application;

[0133] Figure 6d This is a flowchart illustrating a method for generating virtual simulated driving behavior in the upstream perception region during the simulation reproduction stage, as provided in an embodiment of this application.

[0134] Figure 7 This is a flowchart illustrating a method for generating predictive driving simulation behavior provided in an embodiment of this application;

[0135] Figure 8 This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 3 ;

[0136] Figure 9a This is a schematic diagram of a basic traffic map provided in an embodiment of this application;

[0137] Figure 9b This is a line graph illustrating the average vehicle flow rate provided in an embodiment of this application;

[0138] Figure 10a This is a schematic diagram of a scenario in which virtual driving simulation behavior is output in the midstream perception area, as provided in an embodiment of this application;

[0139] Figure 10b This is a flowchart illustrating a method for generating virtual simulated driving behavior in the midstream perception region during the simulation reproduction stage, as provided in an embodiment of this application.

[0140] Figure 11 This application provides an example diagram of a simulation of a main road over time.

[0141] Figure 12 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application;

[0142] Figure 13 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0143] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0144] To facilitate understanding, the following brief explanations are provided for some of the terms:

[0145] Intelligent Vehicle Infrastructure Cooperative Systems (IVICS), or simply vehicle-road cooperative systems, represent a development direction for Intelligent Traffic Systems (ITS). IVICS utilizes advanced wireless communication and next-generation Internet technologies to implement comprehensive, real-time dynamic information interaction between vehicles and infrastructure. Based on the collection and fusion of dynamic traffic information across all times and spaces, it conducts active vehicle safety control and cooperative road management, fully realizing effective collaboration between people, vehicles, and roads. This ensures traffic safety, improves traffic efficiency, and ultimately forms a safe, efficient, and environmentally friendly road traffic system. In the embodiments of this application, the intelligent vehicle-road cooperative system can be used to determine the accurate reproduction simulation and accurate prediction simulation of simulated main roads.

[0146] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0147] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0148] With the research and advancement of artificial intelligence (AI) technology, AI is being researched and applied in various fields, such as smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, autonomous driving, drones, robots, smart healthcare, and smart customer service. It is believed that with further technological development, AI will be applied in even more fields and play an increasingly important role. In this embodiment, AI can be used to generate an autonomous driving model, which represents a comprehensive algorithm module with decision-making, planning, and control execution functions.

[0149] Digital twins are virtual entities created digitally to represent physical entities. They utilize historical data, real-time data, and algorithmic models to simulate, verify, predict, and control the entire lifecycle of a physical entity. Digital twins can establish virtual parallel worlds for roads, mapping the environment, vehicles, and events of the physical road world in real time and completely. Through sensor data distributed along the road, they enable comprehensive perception and dynamic monitoring, forming a precise information representation and mapping of the virtual road to the physical road in terms of information dimensions. This allows management personnel to grasp the overall road condition even when not physically present, solving problems such as difficulty in full-segment detection, delayed event discovery, and difficulty in event review. It not only possesses simulation capabilities but should also have predictive and control capabilities. In the embodiments of this application, digital twins can be used to generate simulated main roads and their corresponding simulation environments.

[0150] In areas covered by sensors, information collected by multi-dimensional traffic facilities such as video and radar is automatically fused. Through target fusion algorithms, the original, discontinuous target information acquired by various sensors corroborates and complements each other, forming a basically complete target attribute information. This target attribute information can be used as perception data in the embodiments of this application, thus enabling accurate depiction of vehicle trajectories on main roads. For example, by establishing relationships between radar-detected targets and video-recognized targets through map correlation, real-time detected targets are overlaid on high-precision maps to achieve the docking of physical and virtual spaces, completing holographic perception of digital mapping. Furthermore, main roads can be simulated in real time in a driving simulation system, and simulation deductions can be performed on this basis to describe, diagnose, predict, and make decisions on core business aspects such as traffic hazards, traffic incidents, and traffic congestion, achieving real-time and efficient intelligent analysis and proactive control, ultimately realizing closed-loop control. This enables refined, intelligent, standardized, and professional management of main roads, laying a solid foundation for traffic governance.

[0151] Please see Figure 1 , Figure 1 This is a schematic diagram of a system architecture provided in an embodiment of this application. For example... Figure 1 As shown, the system may include a business server 100 and a terminal device cluster. The terminal device cluster may include one or more terminal devices; this application does not limit the number of terminal devices. Figure 1 As shown, the terminal device cluster may include terminal device 200a, terminal device 200b, terminal device 200c, ..., terminal device 200n.

[0152] The terminal devices in the cluster can have communication connections with each other. For example, there can be a communication connection between terminal devices 200a and 200b, and between terminal devices 200a and 200c. Simultaneously, any terminal device in the cluster can have a communication connection with the service server 100, for example, terminal device 200a can have a communication connection with the service server 100. The communication connection method is not limited; it can be established directly or indirectly via wired communication, wireless communication, or other methods. This application does not impose any restrictions on this method.

[0153] It should be understood that, such as Figure 1 Each terminal device in the terminal device cluster shown can have an application client installed. When the application client runs on each terminal device, it can interact with the aforementioned... Figure 1 The business server 100 shown interacts with the data, i.e., the communication connection mentioned above. The application client can be a video application, live streaming application, social application, instant messaging application, game application, navigation application, map application, browser, or other application client with the function of loading the simulated main road. This application client can be a standalone client or an embedded sub-client integrated into another client (e.g., a social client, an educational client, or a multimedia client), and there is no limitation here.

[0154] Taking a navigation application as an example, the business server 100 can be a collection of multiple servers, including the backend server corresponding to the navigation application and the data processing server. Therefore, each terminal device can transmit data with the business server 100 through the application client corresponding to the navigation application. For example, each terminal device can upload a road vehicle prediction request for the simulated main road to the business server 100 through the application client of the navigation application. Then, the business server 100 can perform vehicle prediction processing on the simulated main road according to the road vehicle prediction request, obtain the predicted simulated driving behavior of the virtual simulated vehicle, and return the predicted simulated driving behavior of the virtual simulated vehicle to the terminal device.

[0155] It is understood that in the specific implementation of this application, data related to user information (such as perception data and historical data of simulated main roads) are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0156] For ease of subsequent understanding and explanation, the embodiments of this application may be... Figure 1In the terminal device cluster shown, one terminal device is selected as the target terminal device, for example, terminal device 200a. When a road vehicle prediction request for the simulated main road is received, terminal device 200a can send the road vehicle prediction request to the business server 100. Based on the road vehicle prediction request, the business server 100 obtains the driving simulation system, generates a simulated main road in the driving simulation system, and determines the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area in the simulated main road. The sensing data can be real road data, such as data collected by sensing devices on the side of a real main road, which may include the driving route of the real vehicle, the driving position, obstacles around the real vehicle (e.g., obstructing vehicles, pedestrians, and other obstacles), and the location of the obstacles. It is understandable that the simulated main road does not have perception data for the entire road segment because the real road data for some road segments has not been collected, processed, and used to generate perception data. Therefore, before the driving simulation system is run, it is necessary to determine which road segments have perception data and which road segments do not. In this embodiment of the application, the road segments with perception data in the simulated main road are called perception coverage areas, and the road segments without perception data are called perception blank areas.

[0157] The driving simulation system comprises two consecutive simulation phases: a simulation reproduction phase and a simulation prediction phase. Since the perception blank area lacks perception data, and the real road segment it maps to may be inhabited by actual vehicles, the business server 100 generates a first virtual simulation vehicle in the perception blank area at the start of the simulation, based on the regional relationship between the perception coverage area and the perception blank area. This first virtual simulation vehicle is generated to accurately reflect the actual traffic conditions in the perception blank area. Furthermore, in the simulation reproduction phase, which occurs later than the start of the simulation, the business server 100 outputs the reproduced simulated driving behavior corresponding to the perception data in the perception coverage area. This reproduced simulated driving behavior is consistent with the actual driving behavior in the perception data. For example, if the perception data indicates that vehicle a decelerates and moves into the right lane at a certain moment, then in the driving simulation system, for the perception coverage area, the reproduced simulated driving behavior of the vehicle corresponding to vehicle a is deceleration and moving into the right lane.

[0158] Simultaneously, based on the first virtual simulation vehicle, the business server 100 outputs the virtual simulated driving behavior of the second virtual simulation vehicle in the perception blank area, wherein the second virtual simulation vehicle includes the first virtual simulation vehicle. By running the simulation reproduction phase, the business server 100 can determine that the simulated traffic state reproduced in the driving simulation system is highly similar to the actual scene (wherein, the simulated traffic state reproduced within the perception coverage area is consistent with the actual scene). Therefore, it can ensure that the initial simulation data entering the simulation prediction phase is highly similar to the actual scene, thereby determining the accuracy of the predicted simulated driving behavior output by the driving simulation system.

[0159] Furthermore, in the simulation prediction stage, which is later than the simulation reproduction stage, the business server 100 pauses the input of perception data. At this time, the driving simulation system performs prediction simulation on the entire road segment of the simulation main road. Based on the virtual simulation driving behavior and the reproduced simulation driving behavior, the business server 100 outputs the predicted simulation driving behavior of the third virtual simulation vehicle on the simulation main road.

[0160] Subsequently, the service server 100 sends the predicted simulated driving behavior to the terminal device 200a. After receiving the predicted simulated driving behavior sent by the service server 100, the terminal device 200a can display the predicted simulated driving behavior on its corresponding screen. The service server 100 can send the predicted simulated driving behavior to the terminal device 200a in real time; for example, it can send the predicted simulated driving behavior corresponding to each simulation step to the terminal device 200a. Alternatively, the service server 100 can send the predicted simulated driving behavior to the terminal device 200a after the simulation prediction phase ends. Optionally, the service server 100 can also send the predicted simulated driving behavior within an update cycle to the terminal device 200a according to the update cycle. This application embodiment does not limit the method by which the service server 100 sends the predicted simulated driving behavior; it can be set according to the needs of the actual application scenario.

[0161] Optionally, if the terminal device 200a can acquire perception data locally, then the terminal device 200a can create a driving simulation system locally. The subsequent processing is consistent with the process by which the business server 100 generates the predicted simulated driving behavior, so it will not be described in detail here.

[0162] It should be noted that the aforementioned business server 100, terminal device 200a, terminal device 200b, terminal device 200c..., terminal device 200n can all be blockchain nodes in the blockchain network. The data described in the entire text (such as sensing data) can be stored. The storage method can be that the blockchain node generates blocks based on the data and adds the blocks to the blockchain for storage.

[0163] Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. It is primarily used to organize data chronologically and encrypt it into a ledger, making it tamper-proof and forgery-proof. It also allows for data verification, storage, and updating. Essentially, a blockchain is a decentralized database where each node stores an identical blockchain record. The blockchain network can be categorized into core nodes, data nodes, and light nodes. These three types of nodes together constitute the blockchain network. Core nodes are responsible for the consensus of the entire blockchain network; in other words, they are the consensus nodes. The process of writing transaction data into the ledger in a blockchain network can be as follows: data nodes or light nodes in the blockchain network obtain transaction data and pass it through the network (i.e., nodes pass it like a relay) until the consensus node receives the transaction data. The consensus node then packages the transaction data into a block, performs consensus on the block, and writes the transaction data into the ledger after consensus is achieved. Here, we take the example of sensing data to illustrate transaction data. After reaching a consensus on the transaction data, the business server 100 (blockchain node) generates a block based on the transaction data and stores the block in the blockchain network. As for reading the transaction data (i.e., sensing data), the blockchain node can obtain the block containing the transaction data in the blockchain network, and further retrieve the transaction data from the block.

[0164] It is understood that the methods provided in this application embodiment can be executed by computer devices, including but not limited to terminal devices or business servers. The business server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud databases, cloud services, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Terminal devices include but are not limited to mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle terminals, and aircraft. The terminal devices and business servers can be directly or indirectly connected via wired or wireless means, and this application embodiment does not impose any limitations on this connection.

[0165] Further, please see Figure 2 , Figure 2 This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 1This application's embodiments can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, and assisted driving. This application's embodiments are applicable to business scenarios such as route search, route recommendation, and route navigation for main roads; specific business scenarios will not be listed here. This data processing method can be provided by a business server (e.g., the aforementioned...). Figure 1 The business server 100 shown can execute the service, or it can be executed by a terminal device (e.g., the one described above). Figure 1 The terminal device 200a) shown can execute the method, but it can also be executed interactively by the service server and the terminal device. For ease of understanding, this application embodiment uses the execution by the service server as an example for illustration. Figure 2 As shown, the data processing method may include at least the following steps S101-S104.

[0166] Step S101: In the driving simulation system, determine the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area in the simulated main road.

[0167] Specifically, the method proposed in this application embodiment can be embedded in a driving simulation system to perform spatiotemporal simulation of traffic vehicles on the main roads of a digital twin system. Specifically, before the simulation runs, the method in this application embodiment (see step S102) is used to set the parameters. In each simulation step after the simulation runs, the simulation vehicle in the driving simulation system is spatiotemporally simulated, and the motion of the simulation vehicle is described using the method in this application embodiment (see steps S103-S104).

[0168] In this embodiment, the spatiotemporal simulation has two meanings: "spatiotemporal" refers to the motion simulation of virtual vehicles in the perceptual blank area, and "temporal" refers to the simulation of the operating state of the simulated vehicles over a future period after the perceptual data injection is completed. The combination of "temporal" and "spatiotemporal" allows for the determination of the real-time overall condition of the main road and the prediction of traffic trends over a future period. This enables proactive measures to be taken based on the future traffic situation, such as proactive prevention and control measures, to alleviate impending congestion. Steps S102-S104 below describe the simulation of the main road from both "spatiotemporal" and "temporal" perspectives.

[0169] The object of study in this application embodiment can be a digital twin system of a main road (equivalent to the driving simulation system in this application embodiment), thus requiring the perception and simulation of vehicles traveling on the main road. In reality, the perception devices on the main road side have a certain coverage range, which is limited by the type of perception device (such as millimeter-wave radar, cameras, etc.) and weather conditions. In this application embodiment, the part of the main road effectively covered by the perception device is defined as the perception coverage area, which can ensure that all vehicle information in this area can be collected by the perception device and uploaded as perception data to the driving simulation system for complete information mapping and reproduction. In this application embodiment, the part of the main road not effectively covered by the perception device is defined as the perception blank area.

[0170] The business server determines the main road to be simulated in the driving simulation system, obtains the corresponding simulated main road in the system, and identifies which areas in the simulated main road have sensing data. Areas with sensing data are designated as sensing coverage areas, and the remaining areas in the simulated main road (i.e., areas without sensing data) are designated as sensing blank areas. It is understood that different simulated main roads have different lengths and different roadside sensing devices, thus their corresponding sensing coverage areas and sensing blank areas differ. This application embodiment does not limit the number and length of sensing coverage areas in the simulated main road; these should be determined based on the actual application scenario. Similarly, it does not limit the number and length of sensing blank areas in the simulated main road; these should be determined based on the actual application scenario. This application embodiment only requires that the conditions of at least one sensing blank area and at least one sensing coverage area are met.

[0171] Please see also Figure 3 , Figure 3 This is a schematic diagram of a scene including a sensing coverage area and a sensing blank area, provided in an embodiment of this application. For example... Figure 3As shown, the simulated main road generated by the business server in the driving simulation system includes three perception blank areas and two perception coverage areas. The three perception blank areas are perception blank area 301c, perception blank area 303c, and perception blank area 305c. The two perception coverage areas are perception coverage area 302c and perception coverage area 304c. The simulated vehicles in the three perception blank areas can all be understood as virtual simulated vehicles; for example, simulated vehicle 301d is a virtual simulated vehicle. Before the driving simulation system enters the simulation prediction stage, the simulated vehicles in the two perception coverage areas are generated based on perception data and can therefore be understood as reproduced simulated vehicles; for example, simulated vehicle 302d is a reproduced simulated vehicle. This embodiment of the application demonstrates a simulated main road with three simulated lanes, where dashed lines 301b and 302b are lane dividing lines.

[0172] The simulated arterial road described above is obtained by simulating a real arterial road, such as the arterial road in a digital twin system. In this scenario, the sensing data can be real road data collected by roadside sensing devices on the real arterial road. Furthermore, this application example does not limit the real-time nature of the real road data; it can be real-time real road data or replayed real road data. Optionally, the sensing data can be virtual road data, such as virtual road data set by the business server to predict whether a vehicle collision will occur on a real arterial road. Another feasible solution is that the simulated arterial road can also be a virtual arterial road; in this scenario, the sensing data is virtual road data. As mentioned above, this application embodiment does not limit the source of the simulated arterial road and the sensing data; they can be set according to the needs of the actual application scenario.

[0173] Step S102: At the start of the simulation, based on the regional relationship between the sensing coverage area and the sensing blank area, a first virtual simulation vehicle is generated in the sensing blank area.

[0174] Specifically, in the perception data, the initial perception data of the perception coverage area at the start of the simulation is obtained; based on the initial perception data, the initial reproduction simulation vehicle is generated in the perception coverage area; the initial reproduction simulation vehicle belongs to the reproduction simulation vehicle corresponding to the reproduction simulation driving behavior.

[0175] Specifically, based on the regional relationship between the sensing coverage area and the sensing blank area, the initial traffic state corresponding to the sensing blank area is determined; based on the initial traffic state, the first virtual simulation vehicle is generated in the sensing blank area.

[0176] The total number of sensing coverage areas is M, where M is a positive integer. The specific process of determining the initial traffic state corresponding to the sensing blank area based on the regional relationship between the sensing coverage areas and the sensing blank area may include: if the regional relationship between the M sensing coverage areas and the sensing blank area is a downstream regional relationship, then the second historical data of the sensing blank area is obtained; the downstream regional relationship is used to characterize that the sensing blank area is a downstream region of the M sensing coverage areas; if the second historical data is not an empty set, then the second initial historical data corresponding to the simulation start time is obtained from the second historical data, and the second initial historical data is determined as the initial traffic state corresponding to the sensing blank area; if the second historical data is an empty set, then among the M sensing coverage areas, the sensing coverage area that is connected to the sensing blank area and located upstream of the sensing blank area is obtained as the upstream sensing coverage area, and the initial traffic state is determined based on the upstream sensing coverage area.

[0177] The specific process of determining the initial traffic state based on the upstream sensing coverage area may include: obtaining the second initial sensing data corresponding to the upstream sensing coverage area from the initial sensing data; if the second initial sensing data meets the state setting conditions for the sensing blank area, then determining the initial traffic state based on the second initial sensing data; if the second initial sensing data does not meet the state setting conditions, then determining the initial traffic state based on the target traffic state in the third basic traffic map corresponding to the sensing blank area.

[0178] The process of generating a first virtual simulation vehicle in a perception blank area based on the initial traffic state includes: determining the average vehicle spacing corresponding to the perception blank area based on the initial traffic state; if the regional relationship is a downstream regional relationship, generating a first virtual simulation vehicle in the perception blank area based on the average vehicle spacing corresponding to the perception blank area, the upstream edge of the perception blank area, and the driving direction; the downstream regional relationship is used to characterize that the perception blank area is a downstream area of ​​the perception coverage area.

[0179] Before the driving simulation system runs, for perception blank areas lacking perception data, virtual simulation vehicles need to be initialized at the start of the simulation to avoid "gaps" in the road network without vehicles at the beginning of the simulation. Simultaneously, the business server can obtain the initial perception data of the perception coverage area at the start of the simulation from the perception data, and then generate the initial reproduction simulation vehicle in the perception coverage area based on the initial perception data. It is understood that if there are multiple perception coverage areas, each perception coverage area is processed independently. Please see also... Figure 3 ,set up Figure 3 This is a schematic diagram of the scene at the start of the simulation. The driving simulation system generates a starting replica simulation vehicle within the perception coverage area 302c based on the initial perception data corresponding to that area. For example... Figure 3 The example simulated vehicle is 302d. Based on the initial perception data corresponding to the perception coverage area 304c, the driving simulation system generates an initial replica simulated vehicle within the perception coverage area 304c, for example... Figure 3 The example simulated vehicle is 303d.

[0180] Similarly, if the total number of perceived blank areas is multiple, then each perceived blank area is processed independently, such as... Figure 3 As illustrated, the driving simulation system processes the perception blank areas 301c, 303c, and 305c independently. The business server needs to generate the first virtual simulation vehicle in the perception blank area based on the regional relationship between the perception blank area and the perception coverage area. Please refer to [link to relevant documentation]. Figure 3 The sensing blank area 301c is located at the upstream end of the simulated main road. It can also be understood that the sensing blank area 301c is the upstream area of ​​all sensing coverage areas. Therefore, the regional relationship between the sensing blank area 301c and the sensing coverage area is an upstream region relationship. In order to distinguish it from other sensing blank areas, the sensing blank areas with upstream region relationships in this application embodiment are referred to as sensing upstream regions.

[0181] The sensing blank area 303c is located between the sensing coverage areas 302c and 303c; therefore, the regional relationship between the sensing blank area 303c and the sensing coverage areas is a midstream regional relationship. To distinguish it from other sensing blank areas, this embodiment refers to sensing blank areas with a midstream regional relationship as sensing midstream areas. The sensing blank area 305c is located at the downstream end of the simulated main road; it can also be understood as a downstream area of ​​all sensing coverage areas. Therefore, the regional relationship between the sensing blank area 305c and the sensing coverage areas is a downstream regional relationship. To distinguish it from other sensing blank areas, this embodiment refers to sensing blank areas with a downstream regional relationship as sensing downstream areas.

[0182] The boundary between the sensing coverage area and the sensing upstream area (i.e., the downstream edge of the sensing upstream area and the upstream edge of the sensing coverage area) can be defined as the sensing upper boundary, such as... Figure 3 The upper boundary of perception (301a) is defined as the boundary between the sensing coverage area and the sensing downstream area (i.e., the upstream edge of the sensing downstream area and the downstream edge of the sensing coverage area). The lower boundary of perception can be defined as follows: Figure 3 The lower boundary of perception is 304a. The upstream edge of the midstream region of perception can be defined as the lower boundary of perception, such as... Figure 3 The upstream edge of the mid-perception blank region 303c can be defined as the lower boundary of perception 302a. The downstream edge of the mid-perception region can be defined as the upper boundary of perception, such as... Figure 3The downstream edge of the sensing blank area 303c can be defined as the sensing upper boundary 303a.

[0183] In the simulation of the main road, the upper and lower boundaries of perception are both line segments perpendicular to the lane direction (i.e., the driving direction) in the reference line coordinate system (ST coordinate system). The simulation conducted within the perception coverage area is defined as a reproduction simulation, which is to completely restore the perceived vehicle information to the driving simulation system in real time. For the perception blank areas (including the aforementioned upstream, midstream, and downstream perception areas), spatial virtual simulation is required according to the following method.

[0184] If the regional relationship between the sensing coverage area and the sensing blank area is a downstream regional relationship, the service server obtains the second historical data of the sensing blank area. The origin of the second historical data will not be elaborated in this step; please refer to the following text. Figure 8 The description in the corresponding embodiment. In the scenario where the second historical data is not an empty set, the service server extracts the historical data corresponding to the simulation start time from the second historical data as the second starting historical data. The service server uses the second starting historical data as the starting traffic state corresponding to the perception blank area. The starting traffic state includes the starting vehicle density, the starting vehicle flow, and the starting vehicle speed. Based on the starting vehicle density, the service server can determine the average vehicle spacing D1 corresponding to the perception blank area, that is, the distance between two vehicles. Further, the service server generates a normal distribution N(D1, σ) with the average vehicle spacing D1 as the mean. 2 In this application, the specific distribution and variance are not limited, as long as diversity is guaranteed.

[0185] Please also refer to [further details]. Figure 4a , Figure 4a This is a schematic diagram illustrating a scenario where a first virtual simulation vehicle is generated in a downstream sensing area, as provided in an embodiment of this application. Figure 4a For example, the simulated main road includes a sensing coverage area 301e and a sensing blank area 302e. The initial simulated vehicle in the sensing coverage area 301e is reproduced based on the initial sensing data corresponding to the sensing coverage area 301e, and the initial traffic state of the initial simulated vehicle is also determined based on the initial sensing data corresponding to the sensing coverage area 301e. The service server, at the sensing lower boundary 301f (i.e., the upstream edge) of the sensing blank area 302e (also called the sensing downstream area), begins to search downstream (i.e., in the direction of travel) for a distance di (following a normal distribution N(D1, σ)). 2 The random value of ) is used as the position of the virtual simulation vehicle to be filled in the downstream area of ​​perception, and the vehicle spacing dj (j=i, i+1,…) is generated sequentially to fill the virtual vehicles downstream, such as Figure 4a The example shown shows vehicle spacing d1 to vehicle spacing d8. This application embodiment does not limit the filling order; it can be done in lane order, i.e., filling one lane before filling another, or it can fill the lanes closest to the sensing upper boundary 301f first, such as... Figure 4a The example of three lanes is then traced downstream. When the selected lane spacing dj extends downstream beyond the range of the downstream sensing area (i.e., the downstream edge of the downstream sensing area), the business server determines that the initialization of the downstream sensing area for that lane has been completed.

[0186] If the second historical data set is empty, the service server will define the sensing coverage area that is adjacent to the downstream sensing area and located upstream of the downstream sensing area as the upstream sensing coverage area. Based on the upstream sensing coverage area, the server will determine the starting traffic state corresponding to the downstream sensing area. Figure 3 As shown in the example, the sensing coverage area 304c can serve as the upstream sensing coverage area of ​​the sensing blank area 305c, such as... Figure 4a As shown in the example, the sensing coverage area 301e can be used as the upstream sensing coverage area of ​​the sensing blank area 302e.

[0187] If there are more than one initial simulation vehicle in a simulation lane within the upstream sensing coverage area, the service server determines that the upstream sensing coverage area meets the state setting conditions for the downstream sensing area. If there is only one initial simulation vehicle or no initial simulation vehicle in any of the simulation lanes within the upstream sensing coverage area, the service server determines that the upstream sensing coverage area does not meet the state setting conditions for the downstream sensing area.

[0188] When the upstream sensing coverage area meets the state setting conditions for the downstream sensing area, the service server calculates the vehicle spacing between every two simulated vehicles in the simulated lane, and then determines the average vehicle spacing corresponding to that simulated lane. Based on the average vehicle spacing corresponding to each simulated lane, the service server can determine the average vehicle spacing D2 corresponding to the downstream sensing area, generating a normal distribution N(D2, σ) with D2 as the mean. 2 The business server follows a normal distribution N(D2, σ). 2 The process of filling the downstream region of perception with virtual simulated vehicles is related to the normal distribution N(D1, σ). 2 The process of filling the downstream area of ​​perception with virtual simulation vehicles is the same, so it will not be repeated here. Please refer to the description above.

[0189] If the upstream sensing coverage area does not meet the state setting conditions for the downstream sensing area, and the second historical data of the downstream sensing area is an empty set, then the service server randomly selects a vehicle density under the free-driving state (equivalent to the target traffic state) in the third traffic basic map (equivalent to the third traffic basic map mentioned above) corresponding to the downstream sensing area. Based on this vehicle density, the service server determines the average vehicle spacing D3. The subsequent process is the same as the process by which the service server fills the downstream sensing area with virtual simulated vehicles (i.e., the first virtual simulated vehicles) based on the average vehicle spacing D1, so it will not be described in detail here. Please refer to the description above. This step will not be described in detail about the traffic basic map; please refer to the following text. Figure 8 The corresponding embodiment describes the basic traffic map.

[0190] The processing procedure for this step can be found in [link to previous section]. Figure 4b , Figure 4b This is a flowchart illustrating a method for generating virtual simulated vehicles in the downstream sensing region at the start of simulation, as provided in an embodiment of this application. Figure 4b As shown, the method may include the following steps: Step S1021, the service server determines whether historical data exists in the downstream sensing area. Specifically, in this embodiment, the historical data corresponding to the downstream sensing area is defined as second historical data. If historical data exists in the downstream sensing area, the service server executes step S1022; if historical data does not exist in the downstream sensing area, the service server executes step S1023. Step S1022, the service server determines the initial traffic state based on the historical data and executes step S1026. Step S1023, the service server determines whether the upstream sensing coverage area meets the state setting conditions. If the upstream sensing coverage area meets the state setting conditions, the service server executes step S1024; if the upstream sensing coverage area does not meet the state setting conditions, the service server executes step S1025. Step S1024, the service server determines the initial traffic state based on the upstream sensing coverage area and executes step S1026. Step S1025, the service server determines the initial traffic state based on the third traffic basic map and executes step S1026. In step S1026, the service server populates virtual simulated vehicles downstream based on the initial traffic status. In this embodiment, the virtual simulated vehicle generated at the start of the simulation is defined as the first virtual simulated vehicle.

[0191] This step does not describe scenarios where the regional relationship is upstream or midstream; please refer to the following text for details. Figure 5 The description in the corresponding embodiments and the following text Figure 8 The description in the corresponding embodiments.

[0192] Step S103: During the simulation reproduction phase, which is later than the simulation start time, the reproduced simulated driving behavior corresponding to the perception data is output in the perception coverage area, and the virtual simulated driving behavior of the second virtual simulated vehicle is output in the perception blank area according to the first virtual simulated vehicle; the second virtual simulated vehicle includes the first virtual simulated vehicle.

[0193] Specifically, the regional relationship is a downstream regional relationship; the downstream regional relationship is used to characterize that the perception blank area is a downstream area of ​​the perception coverage area; the replica simulation vehicle that drives to the perception blank area from the replica simulation vehicle in the upstream perception coverage area is identified as the sixth virtual simulation vehicle; the replica simulation vehicle in the upstream perception coverage area is generated through perception data; the upstream perception coverage area refers to the area in the perception coverage area that is connected to the perception blank area and located upstream of the perception blank area; the first virtual simulation vehicle and the sixth virtual simulation vehicle are identified as the second virtual simulation vehicle; based on the autonomous driving model corresponding to the perception blank area and the downstream edge of the perception blank area, the virtual simulation driving behavior of the second virtual simulation vehicle is output in the perception blank area; the downstream edge of the perception blank area is used to instruct the driving simulation system to delete the virtual simulation vehicle that drives to the downstream edge of the perception blank area from the second virtual simulation vehicle.

[0194] After the driving simulation system starts running, the business server inputs perception data into the system. Based on this data, the system outputs a simulated vehicle and its trajectory within the perception coverage area in real time, thus reproducing the simulated driving behavior. It also reproduces the vehicle's type, position, speed, and attitude (direction angle) within the perception coverage area. As the simulation clock advances, perception data is continuously injected into the system, and the system reproduces simulated vehicles exhibiting simulated driving behavior one by one within the perception coverage area.

[0195] To maintain the realism of the simulation in the perceived blank areas, the business server updates the longitudinal and lateral speeds of the virtual simulated vehicle in these areas using the corresponding autonomous driving models (including car-following and lane-changing models) for each area. The business server processes perceived blank areas separately based on their upstream, midstream, and downstream relationships.

[0196] At the start of the simulation, an initial vehicle, such as the first virtual simulation vehicle described in step S102, is present in the downstream perception area. After the driving simulation system starts running, the virtual simulation vehicle in the downstream perception area will simulate the longitudinal and lateral driving behavior of the virtual simulation vehicle in the area through the autonomous driving model (including the car-following model and the lane-changing model) corresponding to the downstream perception area. In order to distinguish it from the autonomous driving model corresponding to other blank perception areas, this embodiment defines the autonomous driving model corresponding to the downstream perception area as the downstream autonomous driving model.

[0197] The downstream sensing region, as the name suggests, has a sensing coverage area upstream of it. Please refer to [link / reference needed]. Figure 4c , Figure 4c This is a schematic diagram illustrating a scenario where virtual simulated driving behavior is output in a downstream sensing area, as provided in an embodiment of this application. Figure 4c As shown, the simulated main road includes a perception coverage area 301e and a perception blank area 302e. Here, the perception blank area 302e can be referred to as the downstream perception area, and the perception coverage area 301e can be referred to as the upstream perception coverage area of ​​the downstream perception area. The reproduced simulated vehicle and its output reproduced simulated driving behavior in the perception coverage area 301e are determined by perception data. The virtual simulated driving behavior of the virtual simulated vehicle in the downstream perception area is determined by the downstream autonomous driving model.

[0198] The replicated simulated vehicle generated in the perception coverage area 301e, after crossing the lower perception boundary 301f and entering the downstream perception area, becomes a virtual simulated vehicle (because there is no perception data). At this time, similar to the first virtual simulated vehicle, the replicated simulated vehicle that crosses the lower perception boundary 301f is also simulated in its longitudinal direction by the car-following model in the downstream autonomous driving model. Similarly, its lateral direction is simulated by the lane-changing model in the downstream autonomous driving model. This application embodiment does not limit the type of downstream autonomous driving model. Obviously, the virtual simulated vehicle in the downstream perception area can include both the virtual simulated vehicle generated at the start of the simulation and the replicated simulated vehicle that enters from the upstream perception coverage area.

[0199] If historical data exists in the downstream perception area (equivalent to the second historical data mentioned above), the business server adjusts the parameters of the initial downstream autonomous driving model based on the second historical data to obtain the downstream autonomous driving model mentioned above. If no historical data exists in the downstream perception area, the business server adjusts the parameters of the initial downstream autonomous driving model based on the road type corresponding to the perception blank area (e.g., highways and urban arterial roads) to obtain the downstream autonomous driving model mentioned above.

[0200] To ensure the accuracy of traffic conditions, the business server needs to limit the maximum speed of the virtual simulated vehicle (also known as the first vehicle) closest to the downstream edge in each lane within the downstream sensing area. Please refer to [link / reference]. Figure 4c The first vehicles in the downstream sensing area include virtual simulation vehicles 301g, 302g, and 303g. If historical data (i.e., second historical data) exists in the downstream sensing area, the business server determines the maximum vehicle speed of the first vehicle in each lane of the downstream sensing area based on the average speed corresponding to the simulation reproduction phase (e.g., 5:00 PM to 5:30 PM) from the second historical data. When the first vehicle in the downstream area reaches the downstream edge, it will be removed from the driving simulation system, and the vehicle upstream in the same lane will become the new first vehicle in that lane. For example, when virtual simulation vehicle 303g is removed from the driving simulation system, virtual simulation vehicle 304g becomes the new first vehicle in its simulation lane. If second historical data does not exist, the business server can set the maximum vehicle speed of the first vehicle based on the road type or road speed limit. In addition, the business server determines the maximum vehicle speed of the remaining virtual simulation vehicles in the downstream sensing area, excluding the first vehicle, based on the road type or road speed limit of the downstream sensing area.

[0201] The processing procedure for this step can be found in [link to previous section]. Figure 4d , Figure 4d This is a flowchart illustrating a method for generating virtual simulated driving behavior in the downstream perception area during the simulation reproduction stage, as provided in an embodiment of this application. Figure 4d As shown, the method may include the following steps: Step S1031, the driving simulation system enters the simulation reproduction stage. Step S1032, the business server determines whether historical data, i.e., second historical data, exists in the downstream perception area. If second historical data exists, then proceed to step S1033; if second historical data does not exist, then proceed to step S1034. Step S1033, the business server, based on the second historical data, modifies the initial autonomous driving model (here referring to the downstream initial autonomous driving model). Figure 4d (The following steps are understood similarly) Parameter adjustments are performed, and the maximum vehicle speed of the first vehicle is determined based on the second historical data. Step S1034: The business server adjusts the parameters of the initial autonomous driving model according to the road type; and determines the maximum vehicle speed of the first vehicle according to the road type. Step S1035: The business server determines the maximum vehicle speed of the remaining vehicles other than the first vehicle according to the road type. Step S1036: The business server removes vehicles that have reached the downstream edge, i.e., deletes them.

[0202] This step does not describe scenarios where the regional relationship is upstream or midstream; please refer to the following text for details. Figure 5The description in the corresponding embodiments and the following text Figure 8 The description in the corresponding embodiments.

[0203] Step S104: In the simulation prediction stage, which is later than the simulation reproduction stage, the predicted simulation driving behavior of the third virtual simulation vehicle is output on the simulation main road based on the virtual simulation driving behavior and the reproduced simulation driving behavior.

[0204] Specifically, the virtual simulation vehicle corresponding to the simulated driving behavior that meets the simulation prediction conditions in the virtual simulated driving behavior is identified as the seventh virtual simulation vehicle; the seventh virtual simulation vehicle belongs to the second virtual simulation vehicle; the reproduced simulation vehicle corresponding to the simulated driving behavior that meets the simulation prediction conditions in the reproduced simulated driving behavior is identified as the target reproduced simulation vehicle; the target reproduced simulation vehicle is generated based on perception data; based on the upstream edge of the simulated main road, a second vehicle generation sub-region is determined in the simulated main road; based on the fourth historical data of the perception area to which the second vehicle generation sub-region belongs, an eighth virtual simulation vehicle is generated in the second vehicle generation sub-region; the perception area to which the second vehicle generation sub-region belongs belongs to the perception coverage area or the perception blank area; the seventh virtual simulation vehicle, the target reproduced simulation vehicle, and the eighth virtual simulation vehicle are identified as the third virtual simulation vehicle; based on the autonomous driving model corresponding to the simulated main road and the downstream edge of the simulated main road, the predicted simulated driving behavior of the third virtual simulation vehicle is output on the simulated main road; the downstream edge of the simulated main road is used to instruct the driving simulation system to delete the virtual simulation vehicles in the third virtual simulation vehicle that drive to the downstream edge of the simulated main road.

[0205] The specific process of determining the second vehicle generation sub-region in the simulated main road based on its upstream edge includes: if the upstream edge of the simulated main road belongs to a perception blank area, then the first vehicle generation sub-region corresponding to the perception blank area is determined as the second vehicle generation sub-region of the simulated main road; the first vehicle generation sub-region is used to generate virtual simulated vehicles belonging to the second virtual simulated vehicles during the simulation reproduction stage; if the upstream edge of the simulated main road belongs to a perception coverage area, then the second vehicle generation sub-region is determined in the simulated main road based on its upstream edge and the vehicle generation lines within the simulated main road; the vehicle generation lines within the simulated main road are perpendicular to the driving direction of the simulated main road.

[0206] The specific process of generating an eighth virtual simulation vehicle in the second vehicle generation sub-region based on the fourth historical data of the perception region to which the second vehicle generation sub-region belongs may include: if the fourth historical data of the perception region to which the second vehicle generation sub-region belongs is not an empty set, then predictive historical data is obtained from the fourth historical data, and an eighth virtual simulation vehicle is generated in the second vehicle generation sub-region based on the predictive historical data; if the fourth historical data is an empty set and the perception region to which the second vehicle generation sub-region belongs belongs to a perception blank area, then an eighth virtual vehicle is generated in the second vehicle generation sub-region based on the target traffic state in the fifth basic traffic map corresponding to the perception region to which the second vehicle generation sub-region belongs; if the fourth historical data is an empty set and the perception region to which the second vehicle generation sub-region belongs belongs to a perception coverage area, then an eighth virtual vehicle is generated in the second vehicle generation sub-region based on the perception data.

[0207] Step S104 may further include: identifying the virtual simulation vehicle closest to the downstream edge of the simulated main road among the third virtual simulation vehicles as the second downstream vehicle of the simulated main road; determining the maximum vehicle speed of the second downstream vehicle based on the fifth historical data of the perception area to which the second downstream vehicle belongs; the perception area to which the second downstream vehicle belongs belongs to a perception blank area or a perception coverage area; identifying the virtual simulation vehicles other than the second downstream vehicle among the third virtual simulation vehicles as the third upstream vehicles; determining the maximum vehicle speed of the third upstream vehicle based on the road type corresponding to the perception area to which the third upstream vehicle belongs; and the specific process of outputting the predicted simulated driving behavior of the third virtual simulation vehicle on the simulated main road based on the autonomous driving model corresponding to the simulated main road and the downstream edge of the simulated main road may include: outputting the predicted simulated driving behavior of the third virtual simulation vehicle on the simulated main road based on the autonomous driving model corresponding to the simulated main road, the downstream edge of the simulated main road, the maximum vehicle speed of the third upstream vehicle, and the maximum speed of the second downstream vehicle.

[0208] The simulation prediction conditions in this embodiment are as follows: At the end of the simulation reproduction phase, the position in the virtual simulated driving behavior within the perception blank area is located upstream of the vehicle removal line within the perception blank area. If the perception blank area is an upstream perception area, its corresponding vehicle removal line is the first vehicle removal line described below. If the perception blank area is a midstream perception area, its corresponding vehicle removal line is the second vehicle removal line described below. If the perception blank area is a downstream perception area, its corresponding vehicle removal line is its corresponding downstream edge. At the end of the simulation reproduction phase, the position in the reproduced simulated driving behavior within the perception coverage area is located upstream of the vehicle removal line in the perception blank area that is adjacent to and downstream of the perception coverage area. The simulation prediction condition can also be understood as the vehicle not being removed from the driving simulation system at the end of the simulation reproduction phase.

[0209] In a digital twin driving simulation system, vehicles within the perception coverage area can be digitized and presented in real time within the driving simulation system. This embodiment first initializes the traffic vehicles that may exist in the perception blank area, obtaining a first virtual simulated vehicle. After the simulation begins, the movement of vehicles on the main road is described, ensuring that the simulated vehicle's operation maintains the temporal and spatial continuity of traffic conditions while conforming to traffic operation patterns. Based on this, this embodiment can perform spatiotemporal extrapolation and prediction of the traffic conditions on the simulated main road after the perception data injection stops, thereby achieving the interactive fusion of physical and digital spaces and thus providing better decision-making support.

[0210] As can be seen from the above, in the embodiments of this application, the perception blank areas without perception data are described by simulated vehicles at the simulation start time and in the simulation reproduction stage, which can improve the accuracy of the driving simulation system in reproducing the simulated main road. In addition, the embodiments of this application describe the simulated main road in different simulation stages separately, which can improve the prediction accuracy of the driving simulation system for the simulated main road.

[0211] Please see Figure 5 , Figure 5 This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 2 This method can be implemented by a business server (e.g., the one mentioned above). Figure 1 The business server 100 shown can execute the service, or it can be executed by a terminal device (e.g., the one described above). Figure 1 The terminal device 200a) shown can execute the method, but it can also be executed interactively by the service server and the terminal device. For ease of understanding, this application embodiment uses the execution by the service server as an example for illustration. Figure 5 As shown, the method may include at least the following steps.

[0212] Step S201: In the driving simulation system, determine the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area in the simulated main road; the total number of sensing coverage areas is P, where P is a positive integer.

[0213] Specifically, for example, a real main road 'a' includes a first segment and a second segment. The first segment has sensing devices, which can collect real vehicle information for the corresponding sensing coverage area in the driving simulation system. However, the second segment lacks sensing devices, so real vehicle information cannot be collected, and consequently, the sensing blank area corresponding to the second segment in the driving simulation system has no sensing data.

[0214] Step S202: At the start of the simulation, if the regional relationship between the P sensing coverage areas and the sensing blank areas is an upstream regional relationship or a midstream regional relationship, then among the P sensing coverage areas, the sensing coverage area that is connected to the sensing blank area and located downstream of the sensing blank area is obtained as the first sensing coverage area; the upstream regional relationship is used to characterize that the sensing blank area is an upstream area of ​​the P sensing coverage areas; the midstream regional relationship is used to characterize that the sensing blank area is located between two sensing coverage areas in the P sensing coverage areas.

[0215] For specific scenarios where the regional relationship is a downstream regional relationship, please refer to the above text. Figure 2 The description of step S102 in the corresponding embodiment describes the upstream regional relationship and the midstream regional relationship.

[0216] Please see again. Figure 3 The sensing blank area 301c is located upstream of all sensing coverage areas (including sensing coverage areas 302c and 304c). Therefore, the regional relationship between the sensing blank area 301c and the sensing coverage areas is an upstream regional relationship, and the sensing blank area 301c can be called the sensing upstream area. At this time, the sensing coverage area 302c can be called the first sensing coverage area of ​​the sensing upstream area.

[0217] The perceptual blank area 303c is located between the perceptual coverage areas 302c and 304c. Therefore, the regional relationship between the perceptual blank area 303c and the perceptual coverage areas is a mid-range relationship, and thus the perceptual blank area 303c can be called the perceptual mid-range area. At this time, the perceptual coverage area 304c can be called the first perceptual coverage area of ​​the perceptual mid-range area.

[0218] Step S203: Obtain the first initial sensing data corresponding to the first sensing coverage area from the initial sensing data.

[0219] Specifically, the first sensing coverage area includes A simulated lanes; A is a positive integer; A simulated lanes include simulated lane E f f is a positive integer and f is less than or equal to A; in the first initial perception data, the data located in the simulated lane E is obtained. f The simulated vehicle G in the simulation f Reproduce simulated vehicle G f This belongs to the initial simulation vehicle; if the simulation vehicle G is being reproduced... f If the total number is at least two, then the first initial sensing data is determined to meet the state setting conditions; if the simulated vehicle G is reproduced... f If the total number is less than or equal to 1, then the simulation lane E is determined. fIf A simulated lanes are considered as empty lanes, then the first initial perception data does not meet the state setting conditions.

[0220] Step S204: If the first initial sensing data meets the state setting conditions for the sensing blank area, then the initial traffic state corresponding to the sensing blank area is determined based on the first initial sensing data.

[0221] Specifically, at least two replica simulation vehicles G should be identified. f The distance between any two adjacent simulated vehicles is determined; based on the distance between any two adjacent simulated vehicles, at least two simulated vehicles G are determined. f The average vehicle spacing; at least two simulated vehicles G f The average vehicle spacing is determined as the simulated lane E. f The corresponding average vehicle spacing; based on the average vehicle spacing corresponding to A simulated lanes, determine the vehicle density corresponding to the perception blank area; based on the vehicle density, determine the initial traffic state corresponding to the perception blank area in the first basic traffic map corresponding to the first perception coverage area.

[0222] When the first sensing coverage area meets the state setting conditions for the sensing blank area, the service server calculates the vehicle spacing between every two replicated simulated vehicles in a simulated lane, and then determines the average vehicle spacing corresponding to that simulated lane. Based on the average vehicle spacing corresponding to each simulated lane, the service server can determine the average vehicle spacing D4 corresponding to the first sensing coverage area. Based on the average vehicle spacing D4, the service server can determine the vehicle density corresponding to the first sensing coverage area. Through the pre-set first traffic basic map corresponding to the first sensing coverage area, the service server can determine the traffic state corresponding to the first sensing coverage area. Here, it is assumed that this traffic state is also applicable to the upstream or midstream sensing areas, so the service server determines the traffic state corresponding to the first sensing coverage area as the initial traffic state corresponding to the sensing blank area.

[0223] Furthermore, the business server generates a normal distribution N(D4,σ) with mean D4. 2 This application does not limit the specific distribution and variance in its embodiments, as long as diversity is ensured. Please also refer to... Figure 6a , Figure 6a This is a schematic diagram illustrating a scenario where a first virtual simulation vehicle is generated in the upstream sensing region, as provided in an embodiment of this application. Figure 6aAs shown, the simulated main road includes a sensing blank area 602a and a sensing coverage area 601a. ​​The initial simulated vehicle in the sensing coverage area 601a is reproduced based on the initial sensing data corresponding to the sensing coverage area 601a. ​​The initial traffic state of the initial simulated vehicle is also determined based on the initial sensing data corresponding to the sensing coverage area 601a. ​​The service server starts backtracking upstream (in the opposite direction of travel) by a distance di (following a normal distribution N(D4, σ) from the sensing upper boundary 603a (i.e., the upstream edge) of the sensing blank area 602a (also called the sensing upstream area) upstream (in the opposite direction of travel). 2 The random value of ) is used as the position of the virtual vehicle to be filled in the upstream region of perception, and the vehicle spacing dj (j=i, i+1,…) is generated sequentially and the vehicles are filled upstream, such as Figure 6a The vehicle spacing d9-d14 is illustrated in the example. This application embodiment does not limit the order in which vehicles are filled; it can be done in lane order, i.e., filling one lane before filling another, or it can first fill the lanes closest to the upper boundary 603a of the sensing area, and then trace back upstream. When a selected lane has exceeded the range of the upstream sensing area after tracing back upstream, the service server determines that the initialization work of the upstream sensing area for this lane has been completed.

[0224] Since the process of generating the first virtual simulation vehicle in the upstream sensing region is the same as the process of generating the first virtual simulation vehicle in the midstream sensing region, steps S205-S206 only describe the processing of the upstream sensing region. For the processing of the midstream sensing region, please refer to the description of the upstream sensing region.

[0225] Step S205: If the first initial sensing data does not meet the state setting conditions, then the first historical data of the sensing blank area is obtained, and the initial traffic state is determined based on the first historical data.

[0226] Specifically, if the first historical data is not an empty set, the first initial historical data corresponding to the simulation start time is obtained from the first historical data, and the first initial historical data is determined as the initial traffic state; if the first historical data is an empty set, the initial traffic state is determined according to the target traffic state in the second basic traffic map corresponding to the perceived blank area.

[0227] When the initial sensing coverage area does not meet the state setting conditions for the upstream sensing area, the service server retrieves the historical data corresponding to the simulation start time from the first historical data, using it as the first initial historical data. The service server uses this first initial historical data as the initial traffic state corresponding to the sensing blank area, which includes the initial vehicle density, initial vehicle flow, and initial vehicle speed. Based on the initial vehicle density, the service server can determine the average vehicle spacing D5 corresponding to the sensing blank area, i.e., the distance between two vehicles. Furthermore, the service server generates a normal distribution N(D5, σ) with the average vehicle spacing D5 as the mean. 2 This application does not limit the specific distribution and variance in its embodiments, as long as diversity is ensured. The service server follows a normal distribution N(D5, σ). 2 The process of filling the upstream region of perception with virtual simulated vehicles is related to the normal distribution N(D4, σ). 2 The process of filling the upstream region of perception with virtual simulation vehicles is the same, so it will not be repeated here. Please refer to the description above.

[0228] When the first sensing coverage area does not meet the state setting conditions for the upstream sensing area, and the first historical data is an empty set, the service server randomly selects a vehicle density under the free-driving state (equivalent to the target traffic state mentioned above) in the basic traffic map corresponding to the upstream sensing area. Based on this vehicle density, the service server can determine the average vehicle spacing D6. The subsequent process is the same as the process by which the service server fills the upstream sensing area with virtual simulated vehicles (i.e., the first virtual simulated vehicle) based on the average vehicle spacing D4, so it will not be described in detail here. Please refer to the description above.

[0229] For the upstream sensing area, the business server prioritizes using the traffic state of its downstream sensing coverage area as the initial traffic state of the upstream sensing area at the start of the simulation. This is because when the driving simulation system starts running, the traffic state on both sides at the upper sensing boundary 603a can be kept as consistent as possible, thereby avoiding visual differences.

[0230] Step S206: Generate the first virtual simulation vehicle in the perception blank area based on the initial traffic state.

[0231] Specifically, based on the initial traffic conditions, the average vehicle spacing corresponding to the sensing blank area is determined; if the regional relationship is an upstream regional relationship or a midstream regional relationship, then a first virtual simulated vehicle is generated in the sensing blank area based on the average vehicle spacing corresponding to the sensing blank area, the downstream edge of the sensing blank area, and the direction opposite to the driving direction of the simulated main road; the upstream regional relationship is used to characterize that the sensing blank area is an upstream region of the sensing coverage area; the midstream regional relationship is used to characterize that the total number of sensing coverage areas is at least two, and the sensing blank area is located between two sensing coverage areas in at least two sensing coverage areas.

[0232] The midstream sensing region, as defined, has sensing coverage areas both upstream and downstream. The processing in this region is the same as that in the upstream sensing region, so it will not be elaborated further.

[0233] The processing steps S202-S206 above can be referred to together. Figure 6b , Figure 6b This is a flowchart illustrating a method for generating virtual simulated vehicles in the upstream / midstream perception region at the start of simulation, as provided in an embodiment of this application. Figure 6b As shown, the method may include the following steps: Step S2021, the service server determines whether the first sensing coverage area meets the state setting conditions. If the first sensing coverage area (equivalent to the first initial sensing data) meets the state setting conditions, then step S2022 is executed; if the first sensing coverage area does not meet the state setting conditions, then step S2023 is executed. Step S2022, the service server determines the initial traffic state based on the traffic state of the first sensing coverage area. Step S2023, the service server determines whether there is first historical data in the sensing blank area. If there is first historical data, it is equivalent to the first historical data not being an empty set, then step S2024 is executed; if there is no first historical data, it is equivalent to the first historical data being an empty set, then step S2025 is executed. Step S2024, the service server determines the initial traffic state based on the first historical data; Step S2025, the service server determines the initial traffic state based on the second basic traffic map. Step S2026, the service server fills in virtual simulation vehicles upstream based on the initial traffic state.

[0234] One feasible implementation is that if the midstream sensing area between two sensing coverage areas is less than a certain length threshold, then an initial state setting is not required, and it is assumed that there are no vehicles in between. Here, the threshold is not limited and is related to the location and type of sensing device.

[0235] Step S207: During the simulation reproduction phase, which is later than the simulation start time, the reproduced simulated driving behavior corresponding to the perception data is output in the perception coverage area, and the virtual simulated driving behavior of the second virtual simulated vehicle is output in the perception blank area according to the first virtual simulated vehicle; the second virtual simulated vehicle includes the first virtual simulated vehicle.

[0236] Specifically, the regional relationship is an upstream regional relationship; the upstream regional relationship is used to characterize that the perception blank area is the upstream region of the perception coverage area; based on the upstream edge of the perception blank area and the vehicle generation line within the perception blank area, a first vehicle generation sub-region is determined in the perception blank area; the vehicle generation line is perpendicular to the driving direction of the simulated main road; in the perception data, the downstream perception data corresponding to the second perception coverage area is obtained; the second perception coverage area refers to the area in the perception coverage area that is connected to the perception blank area and located downstream of the perception blank area; based on the downstream perception data, a fourth virtual simulated vehicle is generated in the first vehicle generation sub-region, and the first virtual simulated vehicle and the fourth virtual simulated vehicle are determined as the second virtual simulated vehicle; in the perception blank... Within the region, a first vehicle removal line is determined perpendicular to the driving direction of the simulated main road. The first vehicle removal line is located downstream of the first vehicle generation sub-region, and the distance between the first vehicle removal line and the downstream edge of the perception blank area is less than the distance between the first starting vehicle and the downstream edge of the perception blank area. The first starting vehicle refers to the virtual simulation vehicle among the first virtual simulation vehicles that is closest to the downstream edge of the perception blank area. Based on the autonomous driving model corresponding to the perception blank area and the first vehicle removal line, the virtual simulation driving behavior of the second virtual simulation vehicle is output in the perception blank area. The first vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle among the second virtual simulation vehicles that has driven to the first vehicle removal line.

[0237] The specific process of generating a fourth virtual simulation vehicle in the first vehicle generation sub-region based on downstream sensing data may include: if the downstream sensing data meets the state setting conditions for the sensing blank area, then a fourth virtual simulation vehicle is generated in the first vehicle generation sub-region based on the downstream sensing data; if the downstream sensing data does not meet the state setting conditions, then the third historical data of the sensing blank area is obtained, and a fourth virtual simulation vehicle is generated in the first vehicle generation sub-region based on the third historical data.

[0238] Step S207 may further include: the second sensing coverage area includes Z simulated lanes; Z is a positive integer; the Z simulated lanes include simulated lane X y y is a positive integer and y is less than or equal to Z; in the downstream sensing data, the data located in the simulated lane X is obtained. y The simulated vehicle U in the middle y Reproducing the simulated vehicle Uy This refers to the simulated vehicle corresponding to the simulated driving behavior; if the simulated vehicle U y If the total number is at least two, then the downstream sensing data is determined to meet the state setting conditions; if the simulated vehicle U is reproduced... y If the total number is less than or equal to 1, then the simulation lane X is determined. y If Z simulated lanes are considered as empty lanes, then the downstream sensing data does not meet the state setting conditions.

[0239] Specifically, if the downstream sensing data meets the state setting conditions for the sensing blank area, the process of generating a fourth virtual simulation vehicle in the first vehicle generation sub-region based on the downstream sensing data may include: determining at least two replica simulation vehicles U. y Based on the corresponding reproduction timestamps, the simulation lane X is determined using at least two reproduction timestamps. y Based on the average reproduction time intervals of the Z simulated lanes, the average reproduction time interval corresponding to the first vehicle generation sub-region is determined; at least two simulated vehicles U are determined. y Based on the corresponding vehicle speeds, determine the simulated lane X using at least two vehicle speeds. y Based on the average vehicle speeds of the Z simulated lanes, the average vehicle speed of the first vehicle generation sub-region is determined. Based on the average reproduction time interval of the first vehicle generation sub-region, a fourth virtual simulated vehicle with an initial speed equal to the average vehicle speed of the first vehicle generation sub-region is generated in the first vehicle generation sub-region.

[0240] The specific process of generating a fourth virtual simulation vehicle in the first vehicle generation sub-region based on the third historical data may include: if the third historical data is not an empty set, then reproducing historical data is obtained from the third historical data, and a fourth virtual simulation vehicle is generated in the first vehicle generation sub-region based on the reproduced historical data; if the third historical data is an empty set, then a fourth virtual vehicle is generated in the first vehicle generation sub-region based on the target traffic state in the fourth basic traffic map corresponding to the perceived blank area.

[0241] Step S207 may further include: obtaining an initial autonomous driving model corresponding to the perceived blank area, and obtaining third historical data of the perceived blank area; if the third historical data is not an empty set, adjusting the parameters in the initial autonomous driving model corresponding to the perceived blank area based on the third historical data to obtain an autonomous driving model corresponding to the perceived blank area; if the third historical data is an empty set, adjusting the parameters in the initial autonomous driving model corresponding to the perceived blank area based on the road type corresponding to the perceived blank area to obtain an autonomous driving model corresponding to the perceived blank area.

[0242] Step S207 may further include: identifying the virtual simulation vehicle closest to the downstream edge of the perception blank area among the second virtual simulation vehicles as the downstream vehicle of the perception blank area; determining the maximum vehicle speed of the downstream vehicle based on downstream perception data; identifying the virtual simulation vehicles other than the downstream vehicles among the second virtual simulation vehicles as the first upstream vehicles; determining the maximum vehicle speed of the first upstream vehicle based on the road type corresponding to the perception blank area; and the specific process of outputting the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the autonomous driving model corresponding to the perception blank area and the first vehicle removal line may include: outputting the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the autonomous driving model corresponding to the perception blank area, the first vehicle removal line, the maximum vehicle speed of the first upstream vehicle, and the maximum vehicle speed of the downstream vehicle.

[0243] The specific process of determining the maximum vehicle speed of downstream vehicles based on downstream sensing data may include: if the downstream sensing data indicates that there is a simulated vehicle within the second sensing coverage area, then the simulated vehicle closest to the upstream edge of the second sensing coverage area is identified as the second upstream vehicle; the vehicle speed of the second upstream vehicle is determined as the maximum vehicle speed of the downstream vehicle; the simulated lane to which the second upstream vehicle belongs is the same as the simulated lane to which the downstream vehicle belongs; if the downstream sensing data indicates that there is no simulated vehicle within the second sensing coverage area, and the third historical data of the sensing blank area is not an empty set, then the historical vehicle speed is obtained from the third historical data and determined as the maximum vehicle speed of the downstream vehicle; if the downstream sensing data indicates that there is no simulated vehicle within the second sensing coverage area, and the third historical data of the sensing blank area is an empty set, then the maximum vehicle speed of the downstream vehicle is determined according to the road type corresponding to the sensing blank area.

[0244] The upstream sensing region, as the name suggests, has a sensing coverage area downstream of it. Please refer to [link / reference needed]. Figure 6c , Figure 6c This is a schematic diagram illustrating a scenario where virtual simulated driving behavior is output in the upstream perception region, as provided in an embodiment of this application. Figure 6c As shown, the simulated main road includes a sensing coverage area 601c and a sensing blank area 602c. The sensing blank area 602c can be referred to as the upstream sensing area, and the sensing coverage area 601c can be referred to as the second sensing coverage area of ​​the upstream sensing area. The reproduced simulated vehicle and its output reproduced simulated driving behavior within the sensing coverage area 601c are both determined by sensing data.

[0245] At the start of the simulation, an initial vehicle, such as the first virtual simulation vehicle described in step S206 above, exists in the upstream perception region. After the driving simulation system starts running, the virtual simulation vehicle in the upstream perception region will simulate the longitudinal and lateral driving behavior of the virtual simulation vehicle in the region through the autonomous driving model (including the car-following model and the lane-changing model) corresponding to the upstream perception region. In order to distinguish it from the autonomous driving model corresponding to other blank perception regions, this embodiment defines the autonomous driving model corresponding to the upstream perception region as the upstream autonomous driving model.

[0246] Please see again. Figure 6c After the simulation starts running, the simulated vehicles in the second perception coverage area (i.e., perception coverage area 601c) are reproduced from the perception data. Therefore, virtual simulated vehicles in the upstream perception area (i.e., perception blank area 602c) cannot be introduced into this perception coverage area, otherwise it will cause potential collisions or conflicts between the vehicles. Based on this, this application embodiment proposes a vehicle generation line 605c and a vehicle removal line 604c (equivalent to the first vehicle removal line). The vehicle generation line 605c is located at the upstream edge D7 of the upstream perception area, that is, the distance between the upstream edge of the upstream perception area and the vehicle generation line 605c is D7, and the vehicle generation line 605c is a line segment perpendicular to the lane direction (in the ST coordinate system). The vehicle removal line 604c is located downstream of the upstream perception area, at the upstream edge D8 of the perception coverage area 601c, and is a line segment perpendicular to the lane direction, that is, the distance between the downstream edge of the upstream perception area and the vehicle removal line 604c is D8.

[0247] The road portion between vehicle generation line 605c and the upstream edge of the perception upstream area constitutes the first vehicle generation sub-region. After the simulation starts running, the business server randomly generates virtual simulated vehicles on the lane centerline of the first vehicle generation sub-region. Here, the initial position of the virtual simulated vehicles in the exit area (equivalent to the first vehicle generation sub-region) is not limited, but randomness is maintained to avoid virtual simulated vehicles appearing in the driving simulation system from the same place.

[0248] Please refer to the processing procedure for this step as well. Figure 6d , Figure 6d This is a flowchart illustrating a method for generating virtual simulated driving behavior in the upstream perception region during the simulation reproduction stage, as provided in an embodiment of this application. Figure 6dAs shown, the method includes the following steps. Step S2031: The business server determines whether a simulated vehicle exists in the second perception coverage area. Specifically, if a simulated vehicle exists in the second perception coverage area, step S2032 is executed; if no simulated vehicle exists in the second perception coverage area, step S2033 is executed. Step S2032: The business server determines the traffic state based on the traffic state of the second perception coverage area; prioritizes adjusting the parameters of the initial autonomous driving model based on the third historical data; and determines the maximum vehicle speed of the first vehicle based on the speed of the last vehicle. Specifically, if the second perception coverage area meets the state setting conditions, the initial speed of the traffic flow entering the driving simulation system and the time interval between vehicle departures can be assigned according to the perceived real vehicles. For example, the initial vehicle speed is the same as the average speed of the perceived vehicles, and the departure interval is the same as the average time distance of the perceived vehicles. If historical data exists in the upstream perception area (referred to as the third historical data), the business server uses the third historical data to pre-calibrate the parameters of the upstream autonomous driving initial model so that the model's performance is similar to the historical data of the upstream perception area. Figure 6d The first vehicle in the text is equivalent to the downstream vehicle mentioned above, such as... Figure 6c The example shown includes the first vehicle 601d, the first vehicle 602d, and the first vehicle 603d. Figure 6d The last vehicle in the above is equivalent to the first upstream vehicle mentioned above, such as Figure 6c The last train 601e, the last train 602e, and the last train 603e are shown in the example.

[0249] Step S2033: The business server determines whether third historical data exists in the upstream perception area. Specifically, if it exists, proceed to step S2034; otherwise, proceed to step S2035. Step S2034: The business server determines the traffic state based on the third historical data; adjusts the parameters of the initial autonomous driving model based on the third historical data; and determines the maximum vehicle speed of the first vehicle based on the third historical data. Specifically, the business server assigns values ​​to the initial speed of the traffic flow and the departure interval between two vehicles based on the third historical data. Step S2035: The business server determines the traffic state based on the fourth traffic base map; adjusts the parameters of the initial autonomous driving model based on the road type; and determines the maximum vehicle speed of the first vehicle based on the road type. If the second perception coverage area does not meet the state setting conditions, and the upstream perception area does not have third historical data, then a traffic state in a free driving state (equivalent to the target traffic state) is randomly selected from the traffic base map corresponding to the upstream perception area (referred to as the fourth traffic base map), and the initial speed and departure interval are assigned. If no third historical data exists, the business server can use the default model parameters based on the road type.

[0250] Step S2036: The service server generates a fourth virtual simulation vehicle in the first vehicle generation sub-region based on traffic conditions; and determines the maximum vehicle speed of the remaining vehicles (excluding the first vehicle) based on road type. Step S2037: The service server removes vehicles that have reached the first vehicle removal line. Specifically, since there are replica simulations of real vehicles downstream of the upstream sensing region (i.e., the sensing coverage area), virtual simulation vehicles from the upstream sensing region cannot enter the sensing coverage area. Therefore, when any virtual simulation vehicle (including the initial filling vehicle and the virtual vehicles generated in the exit area after the simulation starts) reaches the first vehicle removal line, it is removed from the driving simulation system to avoid conflict with the replica vehicles in the sensing coverage area. This application's embodiment does not... Figure 6c D7 and D8 are defined, but D7 should ensure that the initial positions of vehicles appearing in the vehicle exit area (equivalent to the first vehicle generation sub-area) have spatial diversity, that is, vehicles do not exit from the same place; D8 should ensure that vehicles in the upstream sensing area are removed from the system before entering the sensing coverage area without causing conflicts. The confirmation of a vehicle being removed from the system when it hits the vehicle exit line or the edge of the map can be set as the front edge of the vehicle or the centroid crossing the line, without any restrictions here.

[0251] Step S208: In the simulation prediction stage, which is later than the simulation reproduction stage, the predicted simulation driving behavior of the third virtual simulation vehicle is output on the simulation main road based on the virtual simulation driving behavior and the reproduced simulation driving behavior.

[0252] Specifically, after entering the simulation prediction stage, the injection of perception data into the driving simulation system is stopped. All vehicles in the driving simulation system are controlled by micro traffic models to control their driving behavior. That is, the longitudinal and lateral driving behaviors of the simulated vehicles in each perception area are controlled by the autonomous driving models corresponding to each perception area.

[0253] Please see also Figure 7 , Figure 7 This is a flowchart illustrating a method for generating predictive driving simulation behavior provided in an embodiment of this application. Figure 7As shown, the method may include steps S2081-S2092. This embodiment does not limit the execution order of steps S2082-S2085 or steps S2086-S2092; these can be set according to the actual application scenario. Step S2081: The driving simulation system performs a simulation prediction phase. Step S2082: The business server determines whether fifth historical data exists. The fifth historical data refers to the historical data corresponding to the perception area of ​​the downstream simulated vehicle (i.e., the second downstream vehicle) on the simulation main road. If fifth historical data exists (equivalent to the fifth historical data not being an empty set), then step S2083 is executed; if fifth historical data does not exist (equivalent to the fifth historical data being an empty set), then step S2084 is executed. Step S2083: The business server determines the maximum vehicle speed of the second downstream vehicle based on the fifth historical data. Step S2084: The maximum vehicle speed of the second downstream vehicle is determined based on the road type of the perception area to which the second downstream vehicle belongs. Step S2085: The business server determines the maximum vehicle speed of the third upstream vehicle based on the road type of the perception area to which the third upstream vehicle belongs. Step S2086: Determine whether the upstream edge of the simulated main road is located within the perception coverage area. If the upstream edge is within the perception coverage area, proceed to step S2087; if the upstream edge is not within the perception coverage area, it is located in the upstream perception area of ​​the perception blank area, in which case proceed to step S2088. Step S2087: A new exit lane is set, generating a new exit area. That is, the business server generates a new exit lane (equivalent to a vehicle generation line) within the upstreammost perception coverage area. Step S2088: The exit area from the simulation reproduction stage is reused. For example... Figure 6c The example shown is the vehicle exit area. Step S2089: Determine whether historical data exists in the sensing area to which the vehicle exit area belongs. Figure 7 The exit area in this context is equivalent to the second vehicle generation sub-area mentioned above. If corresponding historical data exists, proceed to step S2090; otherwise, proceed to step S2091. Step S2090: Determine the traffic status based on the historical data. Figure 7 The historical data in the text refers to the fourth historical data described above. Step S2091: If the perception area to which the vehicle exit area belongs is a perception blank area, then the traffic status is determined according to the fifth basic traffic map. Step S2092: Based on the traffic status, an eighth virtual simulation vehicle is generated in the vehicle exit area, and vehicles that have traveled to the downstream edge of the simulated main road are removed.

[0254] As can be seen from the above, in the embodiments of this application, the perception blank areas without perception data are described by simulated vehicles at the simulation start time and in the simulation reproduction stage, which can improve the accuracy of the driving simulation system in reproducing the simulated main road; in addition, the embodiments of this application describe the simulated main road in different simulation stages separately, which can improve the prediction accuracy of the driving simulation system for the simulated main road.

[0255] Please see Figure 8 , Figure 8 This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 3 This method can be implemented by a business server (e.g., the one mentioned above). Figure 1 The business server 100 shown can execute the service, or it can be executed by a terminal device (e.g., the one described above). Figure 1 The method can be executed by the terminal device 200a), or it can be executed interactively by the service server and the terminal device. For ease of understanding, this application embodiment uses the execution of the method by the terminal device as an example for illustration. Figure 8 As shown, the method may include at least the following steps.

[0256] Step S301: In the driving simulation system, determine the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area in the simulated main road.

[0257] Step S302: At the start of the simulation, based on the regional relationship between the sensing coverage area and the sensing blank area, a first virtual simulation vehicle is generated in the sensing blank area.

[0258] For the specific implementation process of steps S301-S302, please refer to the above text. Figure 2 Steps S101-S102 in the corresponding embodiments will not be described in detail here.

[0259] In traffic flow theory, the basic traffic graph describes the relationship between macroscopic vehicle flow, vehicle density, and vehicle speed within a traffic network. Please also refer to... Figure 9a , Figure 9a This is a schematic diagram of a basic traffic map provided in an embodiment of this application. The horizontal axis of the basic traffic map represents vehicle density, and the vertical axis represents vehicle flow. The basic traffic map can be approximated as two straight line segments forming a triangle with the horizontal axis. Each point on the straight line segment represents a traffic state. The first straight line segment describes the free-flowing state of vehicles, referred to as the target traffic state in this embodiment, and its slope is the free-flow vehicle speed V. max , Figure 9a Taking 80 km / h as an example, as vehicle density increases from 0 to the critical density K... cr ( Figure 9aTaking 25 vehicles / km as an example, during the process, the free-flow vehicle speed remains constant, the traffic capacity (i.e., flow rate) gradually increases, and reaches the critical density K. cr When the maximum throughput capacity Q is reached max ( Figure 9a (Taking 2000 vehicles / hour as an example). As the number of vehicles continues to increase, causing the density to rise further, the vehicle speed gradually decreases, leading to congestion, and the traffic capacity also decreases, as shown in the second straight segment. When the vehicle density increases to the congestion density K... jam ( Figure 9a When the traffic flow reaches 140 vehicles / km (for example), the traffic flow enters a completely congested and stopped state, with both vehicle speed and traffic capacity dropping to 0.

[0260] Congestion density K in a basic traffic map jam It depends only on the distance between the front of each vehicle when traffic is completely congested. Maximum capacity Q max Free-flow velocity and other parameters are road type-related and can be obtained through parameter calibration or by consulting relevant specifications. These three parameters uniquely determine two straight line segments in the basic traffic map. The method for obtaining the basic traffic map is not limited here; it can be defined by specifying the above parameters in the scenario file, or different default basic traffic maps can be set within the simulation system to define the basic traffic attributes of different road types.

[0261] Given a vehicle density, the service server can uniquely determine the traffic state of a vehicle based on the basic traffic map, thereby determining the speed of the traffic flow. The distance D between vehicle centroids is inversely related to the traffic flow density K; therefore, when the traffic flow density K is given, the service server can calculate the initial vehicle spacing D and speed V.

[0262] In this embodiment, the basic traffic maps corresponding to different sensing areas are independent. For example, the basic traffic map corresponding to the second sensing coverage area is independent of the basic traffic map corresponding to the third sensing coverage area. Therefore, the basic traffic map corresponding to the second sensing coverage area can be the same as the basic traffic map corresponding to the third sensing coverage area, or it can be different from the basic traffic map corresponding to the third sensing coverage area. Similarly, the basic traffic map corresponding to the upstream sensing area is independent of the basic traffic map corresponding to the downstream sensing area. Therefore, the basic traffic map corresponding to the upstream sensing area can be the same as the basic traffic map corresponding to the downstream sensing area, or it can be different from the basic traffic map corresponding to the downstream sensing area.

[0263] Before the simulation runs, the administrator can set the free-flow vehicle speed V. max Blockage density K jam Critical density K cr Maximum throughput Qmax Default parameters can be used, or default values ​​can be used to generate a basic traffic map, which can also be called a macro basic map.

[0264] When virtual simulated vehicles are needed, the business server can calculate the traffic density K by reciprocal of the vehicle spacing D. Therefore, the traffic state (flow, speed, density, etc.) can be uniquely determined from the basic traffic map, thereby obtaining the initial speed and vehicle-to-vehicle time distance of the virtual simulated vehicle. The relationship between the vehicle spacing D and the traffic density K (i.e., vehicle density) can be found in the following formula (1), and the initial speed V of the vehicle can be determined by formula (2).

[0265] D = 1 / K(1)

[0266]

[0267] On real-world roads, such as highways, the coverage of sensing devices may be limited, meaning real-time vehicle trajectory data may not cover all highway sections requiring simulation. For sections without sensing device coverage, the first step is to determine if historical traffic conditions described by historical data exist. Historical data can include aggregated data such as average vehicle flow, average vehicle density, and average vehicle speed collected by checkpoint devices; please refer to [link to relevant documentation]. Figure 9b , Figure 9b This is a line graph illustrating the average vehicle traffic flow provided in an embodiment of this application. Figure 9b The display shows a line graph of the 24-hour average traffic flow, with time granularity ranging from minutes, 5 minutes, 10 minutes, 15 minutes, 30 minutes to hours. Depending on the data source classification, similar data curves for vehicle density and average speed can also be displayed. After the simulation runs, the business server can use historical data to adjust certain settings for the current simulation.

[0268] In the embodiments of this application, the historical data corresponding to different sensing areas are independent. For example, the historical data corresponding to the second sensing coverage area is independent of the historical data corresponding to the third sensing coverage area. Therefore, the historical data corresponding to the second sensing coverage area can be the same as the historical data corresponding to the third sensing coverage area; the historical data corresponding to the second sensing coverage area can also be different from the historical data corresponding to the third sensing coverage area. In some application scenarios, there may be historical data corresponding to the second sensing coverage area, but there may be no historical data corresponding to the third sensing coverage area.

[0269] If the perceived blank area is too long, it can be divided into several smaller segments of equal length. In this case, historical data is considered separately for each of these smaller segments.

[0270] If the historical data is a time-varying curve that changes over time (e.g.) Figure 9bAs shown in the figure, the values ​​can also be set according to the simulation time. For example, if there is historical data such as the average vehicle flow, average vehicle density, and average vehicle speed every 15 minutes during the simulation, the simulation time can be set accordingly. That is, the parameters that need to be set will be adjusted every 15 minutes to be as consistent as possible with the historical data.

[0271] Step S303: During the simulation reproduction phase, which is later than the simulation start time, the reproduced simulated driving behavior corresponding to the perception data is output in the perception coverage area. If the regional relationship is a mid-range regional relationship, the reproduced simulated vehicle that drives to the perception blank area in the reproduced simulated vehicle in the third perception coverage area is determined as the fifth virtual simulated vehicle. The mid-range regional relationship is used to characterize that the total number of perception coverage areas is at least two, and the perception blank area is located between two perception coverage areas in at least two perception coverage areas. The reproduced simulated vehicle in the third perception coverage area is generated by the perception data. The third perception coverage area refers to the area in at least two perception coverage areas that is connected to the perception blank area and located downstream of the perception blank area.

[0272] Step S304: The first virtual simulation vehicle and the fifth virtual simulation vehicle are identified as the second virtual simulation vehicle.

[0273] Specifically, this will be described in conjunction with steps S303 and S304. The midstream sensing region, as the name suggests, has a sensing coverage area both upstream and downstream. Please refer to [link / reference needed]. Figure 10a , Figure 10a This is a schematic diagram illustrating a scenario where virtual driving simulation behavior is output within the midstream perception region, as provided in an embodiment of this application. Figure 10a As shown, the simulated main road includes a sensing coverage area 901b, a sensing blank area 902b, and a sensing coverage area 903b. Here, the sensing blank area 902b can be referred to as the mid-range sensing area, and the sensing coverage area 903b can be referred to as the third sensing coverage area of ​​the mid-range sensing area. The simulated vehicle and its output simulated driving behavior in sensing coverage area 901b are determined by the sensing data corresponding to sensing coverage area 901b. Similarly, the simulated vehicle and its output simulated driving behavior in sensing coverage area 903b are determined by the sensing data corresponding to sensing coverage area 903b.

[0274] After the simulation starts running, the upstream of the midstream sensing region receives signals from the third sensing coverage area (e.g., Figure 10aVehicles entering the example perception coverage area 901b), i.e., simulated vehicles coming from the third perception coverage area, after crossing the lower perception boundary 901a and entering the mid-perception area, are all simulated by the autonomous driving model corresponding to the mid-perception area, along with the vehicles that filled the mid-perception area at the start of the simulation, in terms of longitudinal and lateral driving behavior. To distinguish them from the autonomous driving models corresponding to other perception blank areas, this embodiment defines the autonomous driving model corresponding to the mid-perception area as the mid-perception autonomous driving model. There is no limitation on the mid-perception autonomous driving model here. If historical data exists in the mid-perception area, the business server calibrates the parameters of the initial mid-perception autonomous driving model in advance using the historical data, so that the performance of the mid-perception autonomous driving model is similar to the historical data of the mid-perception area. If no historical data exists, the business server can use the default model parameters based on the road type corresponding to the mid-perception area.

[0275] The first vehicle in the midstream perception region refers to the simulated vehicle closest to the downstream edge of the midstream perception region. Please refer to [link to previous section]. Figure 10a The first vehicle in the midstream perception area can include simulated vehicles 901e, 902e, and 903e. The service server determines the speed of the last vehicle in the same lane as the maximum speed of the first vehicle. The last vehicle in the midstream perception area can include 901d (in the same lane as simulated vehicle 901e), 902d (in the same lane as simulated vehicle 902e), and 903d (in the same lane as simulated vehicle 903e). The last vehicle in the midstream perception area refers to the simulated vehicle closest to the upstream edge (equivalent to the upper perception boundary 902a) in the third perception coverage area. Clearly, the speed of the first vehicle in the midstream perception area will not exceed the speed of the last vehicle in the third perception coverage area, thus ensuring the continuity of traffic conditions.

[0276] When the first vehicle in the mid-range perception area reaches the exit line 903c (equivalent to the second vehicle removal line), it will be removed from the driving simulation system, and the vehicle upstream in the same lane will become the new first vehicle in that lane. If there are no vehicles in the third perception coverage area, the service server first checks if historical data exists in the mid-range perception area. If it does, the average speed from the historical data is assigned to the first vehicle in each lane of the mid-range perception area. If there are neither vehicles in the third perception coverage area nor historical data in the mid-range perception area, the service server can set the maximum speed of the first vehicle in the mid-range perception area based on the road type or speed limit corresponding to the mid-range perception area. Furthermore, the maximum speed of the remaining vehicles in the mid-range perception area, excluding the first vehicle, can be set based on the road type or speed limit corresponding to the blank perception area.

[0277] Step S305: In the perception blank area, determine a second vehicle removal line perpendicular to the driving direction of the simulated main road; the distance between the second vehicle removal line and the downstream edge of the perception blank area is less than the distance between the second starting vehicle and the downstream edge of the perception blank area; the second starting vehicle refers to the virtual simulation vehicle in the first virtual simulation vehicle that is closest to the downstream edge of the perception blank area.

[0278] Specifically, to ensure no conflict with vehicles in the third perception coverage area, a vehicle removal line 903c, i.e., a second vehicle removal line, should be set downstream of the midstream perception area. This ensures that simulated vehicles in the midstream perception area are removed from the driving simulation system upon encountering the removal line 903c, thus avoiding collisions or conflicts with replicated vehicles in the third perception coverage area. The width of D9 is not limited here, but D9 should guarantee that vehicles in the midstream perception area are removed from the system before entering the perception coverage area 903b, without causing conflict.

[0279] Step S306: Based on the autonomous driving model corresponding to the perception blank area and the second vehicle removal line, output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area; the second vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle that drives to the second vehicle removal line.

[0280] For the processing steps S303-S306, please refer to [link / reference needed]. Figure 10b , Figure 10b This is a flowchart illustrating a method for generating virtual simulated driving behavior in the midstream perception region during the simulation reproduction stage, as provided in an embodiment of this application. Figure 10b As shown, the method includes the following steps. Step S3031: The business server determines whether a simulated vehicle exists in the third perception coverage area. Specifically, if a simulated vehicle exists in the third perception coverage area, step S3032 is executed; if no simulated vehicle exists in the third perception coverage area, step S3033 is executed. Step S3032: The business server prioritizes adjusting the parameters of the initial autonomous driving model based on historical data (here referring to the historical data corresponding to the midstream perception area); and determines the maximum vehicle speed of the first vehicle based on the speed of the last vehicle. Specifically, the business server prioritizes calibrating the parameters of the midstream autonomous driving initial model based on the historical data corresponding to the midstream perception area, so that the model's performance is similar to the historical data of the midstream perception area. Figure 10b The first car in the middle, such as Figure 10a The example shown includes the lead vehicle 901e, lead vehicle 902e, and lead vehicle 903e. Figure 10b The last train in the middle, such as Figure 10a The last train 901d, the last train 902d, and the last train 903d are shown in the example.

[0281] Step S3033: The business server determines whether historical data exists in the perception midstream area. Specifically, if it exists, proceed to step S3034; otherwise, proceed to step S3035. Step S3034: The business server adjusts the parameters of the initial autonomous driving model based on the historical data; and determines the maximum vehicle speed of the first vehicle based on the historical data. Step S3035: The business server adjusts the parameters of the initial autonomous driving model based on the road type corresponding to the perception midstream area; and determines the maximum vehicle speed of the first vehicle based on the road type. Step S3036: The business server determines the maximum vehicle speed of the remaining vehicles (excluding the first vehicle) based on the road type. Step S3037: The business server removes the vehicles that have reached the second vehicle removal line.

[0282] Step S307: During the simulation reproduction phase, which is later than the simulation start time, the reproduced simulated driving behavior corresponding to the perception data is output in the perception coverage area, and the virtual simulated driving behavior of the second virtual simulated vehicle is output in the perception blank area according to the first virtual simulated vehicle; the second virtual simulated vehicle includes the first virtual simulated vehicle.

[0283] Step S308: In the simulation prediction stage, which is later than the simulation reproduction stage, the predicted simulation driving behavior of the third virtual simulation vehicle is output on the simulation main road based on the virtual simulation driving behavior and the reproduced simulation driving behavior.

[0284] In a microscopic traffic simulation system, each time the simulation clock advances, the speed and position of the simulated vehicles in the system are updated. Their motion behavior can be described by microscopic driving behavior models such as car following and lane changing. In a digital twin simulation system, from a timeline perspective, the simulation can be divided into the following stages, please refer to [link / reference]. Figure 11 , Figure 11 This is an example diagram illustrating a simulation of a main road over time, provided by an embodiment of this application. For example... Figure 11 As shown, the initial state setup phase: Figure 11 Let T0 represent the simulation start time. At this time, the initial state of the simulation needs to be set, including the real vehicles sensed in the sensing coverage area at the start of the simulation, and the initial simulated vehicles in the sensing blank area, i.e., the first virtual simulated vehicles. When setting the initial state at time T0, when filling vehicles upstream / downstream, the limit should be set to ensure that they do not leave the map area. Alternatively, the maximum number of vehicles that can be filled in each area can be limited. When filling vehicles according to the rules described above, filling vehicles can stop when the maximum number is reached. The vehicle spacing D during filling and the vehicle time interval generated in the exit area can be obtained using random numbers that conform to a distribution. A separate random seed can be defined to ensure the consistency of random numbers in multiple simulations.

[0285] Perception Reproduction Phase: After the digital twin simulation starts running, the perception devices transmit vehicle information (position, speed, attitude, etc.) within the perception coverage area back to the driving simulation system in real time, and reproduce and display it in the driving simulation system. Virtual simulation vehicles in the perception blank areas (including the upstream perception area, the midstream perception area, and the downstream perception area) are also simulated through the corresponding autonomous driving model.

[0286] Simulation and prediction phase: The digital twin simulation system can perform simulations to predict traffic conditions over a period of time in the future. Figure 11 Let T1 represent the start time of the simulation. From this moment on, sensor data stops being injected into the driving simulation system, and the system enters the simulation extrapolation phase. The simulation runs based on the traffic conditions at time T1, predicting traffic conditions over a future period. If T2 represents the end time of the simulation, then the period between T1 and T2 is the prediction and extrapolation phase of the simulation; no real-time sensor data is injected during this phase. All vehicles in the system are simulated according to a pre-defined model.

[0287] It is understood that the embodiments involved in this application, such as Figure 2 , Figure 5 as well as Figure 8 The respective embodiments can be combined to generate new embodiments.

[0288] As can be seen from the above, in the embodiments of this application, the perception blank areas without perception data are described by simulated vehicles at the simulation start time and in the simulation reproduction stage, which can improve the accuracy of the driving simulation system in reproducing the simulated main road; in addition, the embodiments of this application describe the simulated main road in different simulation stages separately, which can improve the prediction accuracy of the driving simulation system for the simulated main road.

[0289] Further, please see Figure 12 , Figure 12 This is a schematic diagram of a data processing apparatus provided in an embodiment of this application. The data processing apparatus 1 described above can be used to execute the corresponding steps in the method provided in the embodiment of this application. For example... Figure 12 As shown, the data processing device 1 may include: a region determination module 11, a vehicle generation module 12, a first output module 13, and a second output module 14.

[0290] The region determination module 11 is used in the driving simulation system to determine the perception coverage area with perception data and the perception blank area that does not overlap with the perception coverage area in the simulated main road.

[0291] The vehicle generation module 12 is used to generate a first virtual simulation vehicle in the perception blank area at the start of the simulation, based on the regional relationship between the perception coverage area and the perception blank area.

[0292] The first output module 13 is used to output the reproduced simulated driving behavior corresponding to the perception data in the perception coverage area during the simulation reproduction stage later than the simulation start time, and to output the virtual simulated driving behavior of the second virtual simulated vehicle in the perception blank area according to the first virtual simulated vehicle; the second virtual simulated vehicle includes the first virtual simulated vehicle.

[0293] The second output module 14 is used to output the predicted simulated driving behavior of the third virtual simulated vehicle on the simulated main road, based on the virtual simulated driving behavior and the reproduced simulated driving behavior, in the simulation prediction stage, which is later than the simulation reproduction stage.

[0294] The specific functional implementations of the region determination module 11, vehicle generation module 12, first output module 13, and second output module 14 can be found above. Figure 2 Steps S101-S104 in the corresponding embodiment will not be described again here.

[0295] Please see again Figure 12 The vehicle generation module 12 may include a first determining unit 121 and a first generating unit 122.

[0296] The first determining unit 121 is used to determine the initial traffic state corresponding to the sensing blank area based on the regional relationship between the sensing coverage area and the sensing blank area.

[0297] The first generation unit 122 is used to generate a first virtual simulation vehicle in the perception blank area according to the initial traffic state;

[0298] Among them, the vehicle generation module 12 is also used to obtain the initial perception data of the perception coverage area at the start of the simulation from the perception data;

[0299] The vehicle generation module 12 is also used to generate an initial reproduction simulation vehicle in the perception coverage area based on the initial perception data; the initial reproduction simulation vehicle belongs to the reproduction simulation vehicle corresponding to the reproduction simulation driving behavior.

[0300] The specific functional implementation of the vehicle generation module 12, the first determining unit 121, and the first generation unit 122 can be found in the above description. Figure 2 Step S102 in the corresponding embodiment will not be described again here.

[0301] Please see again Figure 12 The total number of sensing coverage areas is P, where P is a positive integer;

[0302] The first determining unit 121 may include: a first obtaining subunit 1211, a second obtaining subunit 1212, a first determining subunit 1213, and a second determining subunit 1214.

[0303] The first acquisition subunit 1211 is used to acquire, within the P sensing coverage areas, the sensing coverage area that is connected to the sensing blank area and located downstream of the sensing blank area, as the first sensing coverage area if the regional relationship between the P sensing coverage areas and the sensing blank area is an upstream regional relationship or a midstream regional relationship; the upstream regional relationship is used to characterize that the sensing blank area is an upstream region of the P sensing coverage areas; the midstream regional relationship is used to characterize that the sensing blank area is located between two sensing coverage areas in the P sensing coverage areas.

[0304] The second acquisition subunit 1212 is used to acquire the first initial sensing data corresponding to the first sensing coverage area from the initial sensing data;

[0305] The first determining subunit 1213 is used to determine the initial traffic state corresponding to the sensing blank area based on the first initial sensing data if the first initial sensing data meets the state setting conditions for the sensing blank area.

[0306] The second determining subunit 1214 is used to obtain the first historical data of the sensing blank area if the first initial sensing data does not meet the state setting conditions, and determine the initial traffic state based on the first historical data.

[0307] The specific functional implementation methods of the first acquisition subunit 1211, the second acquisition subunit 1212, the first determination subunit 1213, and the second determination subunit 1214 can be found above. Figure 5 Steps S202-S205 in the corresponding embodiment will not be described again here.

[0308] Please see again Figure 12 The first sensing coverage area includes A simulated lanes; A is a positive integer; A simulated lanes include simulated lane E f f is a positive integer and f is less than or equal to A;

[0309] The first determining unit 121 may further include: a third obtaining subunit 1215 and a third determining subunit 1216.

[0310] The third acquisition subunit 1215 is used to acquire the location of the simulated lane E from the first initial perception data. f The simulated vehicle G in the simulation f Reproduce simulated vehicle G f This belongs to the initial reproduction simulation vehicle;

[0311] The third determining subunit 1216 is used to reproduce the simulated vehicle G. f If the total number is at least two, then the first initial sensing data is determined to meet the state setting conditions;

[0312] The third determining subunit 1216 is also used to reproduce the simulated vehicle G. f If the total number is less than or equal to 1, then the simulation lane E is determined. f If A simulated lanes are considered as empty lanes, then the first initial perception data does not meet the state setting conditions.

[0313] The specific functional implementation methods of the third acquisition subunit 1215 and the third determination subunit 1216 can be found in the above description. Figure 5 Step S203 in the corresponding embodiment will not be described again here.

[0314] Please see again Figure 12 The first determining subunit 1213 is specifically used to determine at least two reproducible simulation vehicles G. f The distance between any two adjacent simulated vehicles in the simulation.

[0315] The first determining subunit 1213 is further specifically used to determine at least two replica simulation vehicles G based on the vehicle spacing between every two adjacent replica simulation vehicles. f The average vehicle spacing;

[0316] The first determining subunit 1213 is also specifically used to integrate at least two simulated vehicles G. f The average vehicle spacing is determined as the simulated lane E. f The corresponding average vehicle spacing;

[0317] The first determining subunit 1213 is also specifically used to determine the vehicle density corresponding to the perception blank area based on the average vehicle spacing corresponding to A simulated lanes.

[0318] The first determining subunit 1213 is also specifically used to determine the initial traffic state corresponding to the sensing blank area in the first basic traffic map corresponding to the first sensing coverage area based on the vehicle density.

[0319] The specific functional implementation of the first determining subunit 1213 can be found in the above description. Figure 5 Step S204 in the corresponding embodiment will not be described again here.

[0320] Please see again Figure 12 The second determining subunit 1214 is specifically used to obtain the first starting historical data corresponding to the simulation start time from the first historical data if the first historical data is not an empty set, and to determine the first starting historical data as the starting traffic state.

[0321] The second determining subunit 1214 is also specifically used to determine the initial traffic state based on the target traffic state in the second basic traffic map corresponding to the perceived blank area if the first historical data is an empty set.

[0322] The specific functional implementation of the second determining subunit 1214 can be found in the above description. Figure 5 Step S205 in the corresponding embodiment will not be described again here.

[0323] Please see again Figure 12 The total number of sensing coverage areas is M, where M is a positive integer;

[0324] The first determining unit 121 may include: a fourth obtaining subunit 1217, a fifth obtaining subunit 1218, and a sixth obtaining subunit 1219.

[0325] The fourth acquisition subunit 1217 is used to acquire the second historical data of the sensing blank area if the regional relationship between the M sensing coverage areas and the sensing blank area is a downstream regional relationship; the downstream regional relationship is used to characterize that the sensing blank area is a downstream region of the M sensing coverage areas.

[0326] The fifth acquisition subunit 1218 is used to acquire the second starting historical data corresponding to the simulation start time from the second historical data if the second historical data is not an empty set, and to determine the second starting historical data as the starting traffic state corresponding to the perception blank area.

[0327] The sixth acquisition subunit 1219 is used to acquire, in the M sensing coverage areas, the sensing coverage area that is connected to the sensing blank area and located upstream of the sensing blank area, as the upstream sensing coverage area, if the second historical data is an empty set, and determine the starting traffic state based on the upstream sensing coverage area.

[0328] The specific functional implementation methods of the fourth acquisition subunit 1217, the fifth acquisition subunit 1218, and the sixth acquisition subunit 1219 can be found above. Figure 2 Step S102 in the corresponding embodiment will not be described again here.

[0329] Please see again Figure 12 The sixth acquisition subunit 1219 is specifically used to acquire the second initial sensing data corresponding to the upstream sensing coverage area from the initial sensing data;

[0330] The sixth acquisition subunit 1219 is also specifically used to determine the initial traffic state based on the second initial perception data if the second initial perception data meets the state setting conditions for the perception blank area.

[0331] The sixth acquisition subunit 1219 is also specifically used to determine the initial traffic state based on the target traffic state in the third basic traffic map corresponding to the perceived blank area if the second initial sensing data does not meet the state setting conditions.

[0332] The specific functional implementation of the sixth acquisition subunit 1219 can be found in the above description. Figure 2 Step S102 in the corresponding embodiment will not be described again here.

[0333] Please see again Figure 12 The first generation unit 122 may include: a fourth determining subunit 1221, a first generation subunit 1222, and a second generation subunit 1223.

[0334] The fourth determining subunit 1221 is used to determine the average vehicle spacing corresponding to the sensing blank area based on the initial traffic conditions.

[0335] The first generation subunit 1222 is used to generate a first virtual simulation vehicle in the perception blank area if the regional relationship is an upstream regional relationship or a midstream regional relationship, based on the average vehicle spacing corresponding to the perception blank area, the downstream edge of the perception blank area, and the direction opposite to the driving direction of the simulated main road; the upstream regional relationship is used to characterize that the perception blank area is an upstream area of ​​the perception coverage area; the midstream regional relationship is used to characterize that the total number of perception coverage areas is at least two, and the perception blank area is located between two perception coverage areas in at least two perception coverage areas;

[0336] The second generation subunit 1223 is used to generate a first virtual simulation vehicle in the perception blank area based on the average vehicle spacing corresponding to the perception blank area, the upstream edge of the perception blank area, and the driving direction if the regional relationship is a downstream regional relationship; the downstream regional relationship is used to characterize that the perception blank area is a downstream area of ​​the perception coverage area.

[0337] The specific functional implementation methods of the fourth determining subunit 1221, the first generating subunit 1222, and the second generating subunit 1223 can be found above. Figure 2 Step S102 in the corresponding embodiment will not be described again here.

[0338] Please see again Figure 12 The regional relationship is the upstream regional relationship; the upstream regional relationship is used to characterize the perceived blank area as the upstream region of the perceived coverage area;

[0339] The first output module 13 may include: a second determining unit 131, a first acquiring unit 132, a second generating unit 133, a third determining unit 134, and a first output unit 135.

[0340] The second determining unit 131 is used to determine a first vehicle generation sub-region in the perception blank region based on the upstream edge of the perception blank region and the vehicle generation line in the perception blank region; the vehicle generation line is perpendicular to the driving direction of the simulated main road.

[0341] The first acquisition unit 132 is used to acquire downstream sensing data corresponding to the second sensing coverage area from the sensing data; the second sensing coverage area refers to the area in the sensing coverage area that is connected to the sensing blank area and located downstream of the sensing blank area.

[0342] The second generation unit 133 is used to generate a fourth virtual simulation vehicle in the first vehicle generation sub-region based on downstream perception data, and to determine the first virtual simulation vehicle and the fourth virtual simulation vehicle as the second virtual simulation vehicle.

[0343] The third determining unit 134 is used to determine a first vehicle removal line perpendicular to the driving direction of the simulated main road in the perception blank area; the first vehicle removal line is located in the downstream area of ​​the first vehicle generation sub-area, and the distance between the first vehicle removal line and the downstream edge of the perception blank area is less than the distance between the first starting vehicle and the downstream edge of the perception blank area; the first starting vehicle refers to the virtual simulation vehicle that is closest to the downstream edge of the perception blank area among the first virtual simulation vehicles.

[0344] The first output unit 135 is used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area according to the autonomous driving model corresponding to the perception blank area and the first vehicle removal line; the first vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle in the second virtual simulation vehicle that drives to the first vehicle removal line.

[0345] The specific functional implementation of the second determining unit 131, the first acquiring unit 132, the second generating unit 133, the third determining unit 134, and the first output unit 135 can be found above. Figure 5 Step S207 in the corresponding embodiment will not be described again here.

[0346] Please see again Figure 12 The second generation unit 133 may include a third generation subunit 1331 and a fourth generation subunit 1332.

[0347] The third generation subunit 1331 is used to generate a fourth virtual simulation vehicle in the first vehicle generation subregion based on the downstream sensing data if the downstream sensing data meets the state setting conditions for the sensing blank area.

[0348] The fourth generation subunit 1332 is used to obtain the third historical data of the perception blank area if the downstream perception data does not meet the state setting conditions, and generate the fourth virtual simulation vehicle in the first vehicle generation sub-area based on the third historical data.

[0349] The specific functional implementation methods of the third generation subunit 1331 and the fourth generation subunit 1332 can be found in the above description. Figure 5 Step S207 in the corresponding embodiment will not be described again here.

[0350] Please see again Figure 12 The second sensing coverage area includes Z simulated lanes; Z is a positive integer; the Z simulated lanes include simulated lane X y y is a positive integer and y is less than or equal to Z;

[0351] The second generation unit 133 may further include: a seventh acquisition subunit 1333 and a fifth determination subunit 1334.

[0352] The seventh acquisition subunit 1333 is used to acquire the location of the simulated lane X from the downstream sensing data. y The simulated vehicle U in the middle y Reproducing the simulated vehicle U y These are the simulated vehicles that correspond to the simulated driving behavior.

[0353] The fifth determining subunit 1334 is used to reproduce the simulated vehicle U. y If the total number is at least two, then the downstream sensing data is determined to meet the state setting conditions;

[0354] The fifth determining subunit 1334 is also used to reproduce the simulated vehicle U. y If the total number is less than or equal to 1, then the simulation lane X is determined. y If Z simulated lanes are considered as empty lanes, then the downstream sensing data does not meet the state setting conditions.

[0355] The specific functional implementation methods of the seventh acquisition subunit 1333 and the fifth determination subunit 1334 can be found above. Figure 5 Step S207 in the corresponding embodiment will not be described again here.

[0356] Please see again Figure 12 The third generation subunit 1331 is specifically used to determine at least two replica simulation vehicles U. y Based on the corresponding reproduction timestamps, the simulation lane X is determined using at least two reproduction timestamps. yThe corresponding average reproduction time interval is determined based on the average reproduction time intervals corresponding to the Z simulated lanes, thus determining the average reproduction time interval corresponding to the first vehicle generation sub-region.

[0357] The third generation subunit 1331 is also specifically used to determine at least two replica simulation vehicles U. y Based on the corresponding vehicle speeds, determine the simulated lane X using at least two vehicle speeds. y The average vehicle speed is determined based on the average vehicle speeds of the Z simulated lanes, which determines the average vehicle speed of the first vehicle generation sub-region.

[0358] The third generation sub-unit 1331 is also specifically used to generate a fourth virtual simulation vehicle in the first vehicle generation sub-region with an initial speed equal to the average vehicle speed corresponding to the first vehicle generation sub-region, based on the average reproduction time interval corresponding to the first vehicle generation sub-region.

[0359] The specific functional implementation of the third generating subunit 1331 can be found in the above description. Figure 5 Step S207 in the corresponding embodiment will not be described again here.

[0360] Please see again Figure 12 The fourth generation subunit 1332 is specifically used to obtain reproduced historical data from the third historical data if the third historical data is not an empty set, and to generate a fourth virtual simulation vehicle in the first vehicle generation sub-region based on the reproduced historical data.

[0361] The fourth generation sub-unit 1332 is specifically used to generate a fourth virtual vehicle in the first vehicle generation sub-region if the third historical data is an empty set, based on the target traffic state in the fourth basic traffic map corresponding to the perceived blank area.

[0362] The specific functional implementation of the fourth generating subunit 1332 can be found in the above description. Figure 5 Step S207 in the corresponding embodiment will not be described again here.

[0363] Please see again Figure 12 The first output module 13 may further include: a second acquisition unit 136 and a parameter adjustment unit 137.

[0364] The second acquisition unit 136 is used to acquire the initial autonomous driving model corresponding to the perception blank area and to acquire the third historical data of the perception blank area.

[0365] The parameter adjustment unit 137 is used to adjust the parameters in the initial autonomous driving model corresponding to the perception blank area based on the third historical data if the third historical data is not an empty set, so as to obtain the autonomous driving model corresponding to the perception blank area.

[0366] The parameter adjustment unit 137 is also used to adjust the parameters in the initial autonomous driving model corresponding to the perceived blank area according to the road type corresponding to the perceived blank area if the third historical data is an empty set, so as to obtain the autonomous driving model corresponding to the perceived blank area.

[0367] The specific functional implementation of the second acquisition unit 136 and the parameter adjustment unit 137 can be found in the above description. Figure 5 Step S208 in the corresponding embodiment will not be described again here.

[0368] Please see again Figure 12 The first output module 13 may further include: a fourth determining unit 138, a fifth determining unit 139, a sixth determining unit 240, and a seventh determining unit 241.

[0369] The fourth determining unit 138 is used to determine the virtual simulation vehicle in the second virtual simulation vehicle that is closest to the downstream edge of the perception blank area as the downstream vehicle of the perception blank area.

[0370] The fifth determining unit 139 is used to determine the maximum vehicle speed of the downstream vehicle based on the downstream sensing data.

[0371] The sixth determining unit 240 is used to determine the virtual simulation vehicles other than the downstream vehicles in the second virtual simulation vehicle as the first upstream vehicles;

[0372] The seventh determining unit 241 is used to determine the maximum vehicle speed of the first upstream vehicle based on the road type corresponding to the perceived blank area.

[0373] The first output unit 135 is specifically used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the autonomous driving model corresponding to the perception blank area, the first vehicle removal line, the maximum vehicle speed of the first upstream vehicle, and the maximum vehicle speed of the downstream vehicle.

[0374] The specific functional implementation methods of the fourth determining unit 138, the fifth determining unit 139, the sixth determining unit 240, the seventh determining unit 241, and the first output unit 135 can be found above. Figure 5 Step S208 in the corresponding embodiment will not be described again here.

[0375] Please see again Figure 12 The fifth determining unit 139 may include: a sixth determining subunit 1391, a seventh determining subunit 1392, an eighth obtaining subunit 1393, and an eighth determining subunit 1394.

[0376] The sixth determining subunit 1391 is used to determine the second upstream vehicle as the replica simulation vehicle that is closest to the upstream edge of the second sensing coverage area among the replica simulation vehicles in the second sensing coverage area if the downstream sensing data indicates that there is a replica simulation vehicle in the second sensing coverage area.

[0377] The seventh determining subunit 1392 is used to determine the vehicle speed of the second upstream vehicle as the maximum vehicle speed of the downstream vehicle; the simulation lane to which the second upstream vehicle belongs is the same as the simulation lane to which the downstream vehicle belongs.

[0378] The eighth acquisition subunit 1393 is used to acquire the historical vehicle speed from the third historical data if the downstream sensing data indicates that there is no reproduced simulation vehicle in the second sensing coverage area and the third historical data of the sensing blank area is not an empty set, and to determine the historical vehicle speed as the maximum vehicle speed of the downstream vehicle.

[0379] The eighth determining subunit 1394 is used to determine the maximum vehicle speed of the downstream vehicle based on the road type corresponding to the sensing blank area if the downstream sensing data indicates that there is no reproduced simulated vehicle in the second sensing coverage area and the third historical data of the sensing blank area is an empty set.

[0380] The specific functional implementation methods of the sixth determining subunit 1391, the seventh determining subunit 1392, the eighth obtaining subunit 1393, and the eighth determining subunit 1394 can be found above. Figure 5 Step S208 in the corresponding embodiment will not be described again here.

[0381] Please see again Figure 12 The regional relationship is a midstream regional relationship; the midstream regional relationship is used to characterize that the total number of sensing coverage areas is at least two, and the sensing blank area is located between two sensing coverage areas in the at least two sensing coverage areas;

[0382] The first output module 13 may include an eighth determining unit 242, a ninth determining unit 243, and a second output unit 244.

[0383] The eighth determining unit 242 is used to determine the replica simulation vehicle that drives to the perception blank area among the replica simulation vehicles in the third perception coverage area as the fifth virtual simulation vehicle; the replica simulation vehicle in the third perception coverage area is generated by perception data; the third perception coverage area refers to the area that is connected to the perception blank area and located downstream of the perception blank area among at least two perception coverage areas.

[0384] The eighth determining unit 242 is also used to determine the first virtual simulation vehicle and the fifth virtual simulation vehicle as the second virtual simulation vehicle;

[0385] The ninth determining unit 243 is used to determine a second vehicle removal line perpendicular to the driving direction of the simulated main road in the sensing blank area; the distance between the second vehicle removal line and the downstream edge of the sensing blank area is less than the distance between the second starting vehicle and the downstream edge of the sensing blank area; the second starting vehicle refers to the virtual simulation vehicle in the first virtual simulation vehicle that is closest to the downstream edge of the sensing blank area.

[0386] The second output unit 244 is used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area according to the autonomous driving model corresponding to the perception blank area and the second vehicle removal line; the second vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle that drives to the second vehicle removal line.

[0387] The specific functional implementation methods of the eighth determining unit 242, the ninth determining unit 243, and the second output unit 244 can be found above. Figure 8 Steps S303-S308 in the corresponding embodiment will not be described again here.

[0388] Please see again Figure 12 The regional relationship is the downstream regional relationship; the downstream regional relationship is used to characterize the perceived blank area as the downstream region of the perceived coverage area;

[0389] The first output module 13 may include a tenth determining unit 245 and a third output unit 246.

[0390] The tenth determining unit 245 is used to determine the replica simulation vehicle that drives to the perception blank area among the replica simulation vehicles in the upstream perception coverage area as the sixth virtual simulation vehicle; the replica simulation vehicle in the upstream perception coverage area is generated by perception data; the upstream perception coverage area refers to the area in the perception coverage area that is connected to the perception blank area and located upstream of the perception blank area.

[0391] The tenth determining unit 245 is also used to determine the first virtual simulation vehicle and the sixth virtual simulation vehicle as the second virtual simulation vehicle;

[0392] The third output unit 246 is used to output the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area according to the autonomous driving model corresponding to the perception blank area and the downstream edge of the perception blank area; the downstream edge of the perception blank area is used to instruct the driving simulation system to delete the virtual simulation vehicle in the second virtual simulation vehicle that drives to the downstream edge of the perception blank area.

[0393] The specific functional implementation methods of the tenth determining unit 245 and the third output unit 246 can be found in the above description. Figure 2 Step S104 in the corresponding embodiment will not be described again here.

[0394] Please see again Figure 12 The second output module 14 may include: a vehicle determination unit 141, a target determination unit 142, a third generation unit 143, a fourth generation unit 144, and a fourth output unit 145.

[0395] The vehicle determination unit 141 is used to determine the virtual simulation vehicle corresponding to the simulation driving behavior that meets the simulation prediction conditions in the virtual simulation driving behavior as the seventh virtual simulation vehicle; the seventh virtual simulation vehicle belongs to the second virtual simulation vehicle.

[0396] The target determination unit 142 is used to determine the simulated vehicle corresponding to the simulated driving behavior that meets the simulation prediction conditions in the simulated driving behavior as the target simulated vehicle; the target simulated vehicle is generated based on the perception data;

[0397] The third generation unit 143 is used to determine the second vehicle generation sub-region in the simulated main road based on the upstream edge of the simulated main road.

[0398] The fourth generation unit 144 is used to generate an eighth virtual simulation vehicle in the second vehicle generation sub-region based on the fourth historical data of the perception region to which the second vehicle generation sub-region belongs; the perception region to which the second vehicle generation sub-region belongs belongs to the perception coverage area or the perception blank area.

[0399] The vehicle determination unit 141 is also used to determine the seventh virtual simulation vehicle, the target reproduction simulation vehicle, and the eighth virtual simulation vehicle as the third virtual simulation vehicle;

[0400] The fourth output unit 145 is used to output the predicted simulated driving behavior of the third virtual simulated vehicle on the simulated main road according to the autonomous driving model corresponding to the simulated main road and the downstream edge of the simulated main road; the downstream edge of the simulated main road is used to instruct the driving simulation system to delete the virtual simulated vehicle that drives to the downstream edge of the simulated main road in the third virtual simulated vehicle.

[0401] The specific functional implementation methods of the vehicle determination unit 141, target determination unit 142, third generation unit 143, fourth generation unit 144, and fourth output unit 145 can be found above. Figure 2 Step S104 in the corresponding embodiment will not be described again here.

[0402] Please see again Figure 12 The third generation unit 143 may include: a ninth determining subunit 1431 and a tenth determining subunit 1432.

[0403] The ninth determining sub-unit 1431 is used to determine the first vehicle generation sub-region corresponding to the perception blank area as the second vehicle generation sub-region of the simulation main road if the upstream edge of the simulated main road belongs to the perception blank area; the first vehicle generation sub-region is used to generate virtual simulation vehicles belonging to the second virtual simulation vehicles in the simulation reproduction stage.

[0404] The tenth determining sub-unit 1432 is used to determine a second vehicle generation sub-region in the simulated main road based on the upstream edge of the simulated main road and the vehicle generation line in the simulated main road if the upstream edge of the simulated main road belongs to the sensing coverage area; the vehicle generation line in the simulated main road is perpendicular to the driving direction of the simulated main road.

[0405] The specific functional implementation methods of the ninth determining subunit 1431 and the tenth determining subunit 1432 can be found in the above description. Figure 2 Step S104 in the corresponding embodiment will not be described again here.

[0406] Please see again Figure 12 The fourth generation unit 144 may include: a fifth generation subunit 1441, a sixth generation subunit 1442, and a seventh generation subunit 1443.

[0407] The fifth generation sub-unit 1441 is used to obtain predicted historical data from the fourth historical data if the fourth historical data of the perception area to which the second vehicle generation sub-region belongs is not an empty set, and to generate an eighth virtual simulation vehicle in the second vehicle generation sub-region based on the predicted historical data.

[0408] The sixth generation sub-unit 1442 is used to generate an eighth virtual vehicle in the second vehicle generation sub-region if the fourth historical data is an empty set and the perception region to which the second vehicle generation sub-region belongs is a perception blank region. This is based on the target traffic state in the fifth basic traffic map corresponding to the perception region to which the second vehicle generation sub-region belongs.

[0409] The seventh generation sub-unit 1443 is used to generate an eighth virtual vehicle in the second vehicle generation sub-region based on the perception data if the fourth historical data is an empty set and the perception region to which the second vehicle generation sub-region belongs is a perception coverage region.

[0410] The specific functional implementations of the fifth generation subunit 1441, the sixth generation subunit 1442, and the seventh generation subunit 1443 can be found above. Figure 2 Step S104 in the corresponding embodiment will not be described again here.

[0411] Please see again Figure 12The second output module 14 may further include a downstream determination unit 146 and a speed determination unit 147.

[0412] Downstream determination unit 146 is used to determine the virtual simulation vehicle that is closest to the downstream edge of the simulated main road among the third virtual simulation vehicles as the second downstream vehicle of the simulated main road.

[0413] The speed determination unit 147 is used to determine the maximum vehicle speed of the second downstream vehicle based on the fifth historical data of the sensing area to which the second downstream vehicle belongs; the sensing area to which the second downstream vehicle belongs belongs to a sensing blank area or a sensing coverage area.

[0414] The downstream determination unit 146 is also used to determine the virtual simulation vehicles other than the second downstream vehicles in the third virtual simulation vehicle as the third upstream vehicles.

[0415] The speed determination unit 147 is also used to determine the maximum vehicle speed of the third upstream vehicle based on the road type corresponding to the perception area to which the third upstream vehicle belongs.

[0416] The fourth output unit 145 is specifically used to output the predicted simulated driving behavior of the third virtual simulated vehicle on the simulated main road based on the autonomous driving model corresponding to the simulated main road, the downstream edge of the simulated main road, the maximum vehicle speed of the third upstream vehicle, and the maximum speed of the second downstream vehicle.

[0417] The specific functional implementation methods of the downstream determination unit 146 and the speed determination unit 147 can be found in the above description. Figure 2 Step S104 in the corresponding embodiment will not be described again here.

[0418] As can be seen from the above, in the embodiments of this application, the perception blank areas without perception data are described by simulated vehicles at the simulation start time and in the simulation reproduction stage, which can improve the accuracy of the driving simulation system in reproducing the simulated main road; in addition, the embodiments of this application describe the simulated main road in different simulation stages separately, which can improve the prediction accuracy of the driving simulation system for the simulated main road.

[0419] Further, please see Figure 13 , Figure 13 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Figure 13As shown, the computer device 1000 may include: at least one processor 1001, such as a CPU; at least one network interface 1004; a user interface 1003; a memory 1005; and at least one communication bus 1002. The communication bus 1002 is used to enable communication between these components. In some embodiments, the user interface 1003 may include a display screen and a keyboard, and the network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as at least one disk storage device. Optionally, the memory 1005 may also be at least one storage device located remotely from the aforementioned processor 1001. Figure 13 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a device control application program.

[0420] exist Figure 13 In the computer device 1000 shown, the network interface 1004 provides network communication functionality; the user interface 1003 is mainly used to provide an input interface for the user; and the processor 1001 can be used to call the device control application stored in the memory 1005 to achieve:

[0421] In the driving simulation system, the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area are determined in the simulated main road.

[0422] At the start of the simulation, based on the regional relationship between the sensing coverage area and the sensing blank area, the first virtual simulation vehicle is generated in the sensing blank area.

[0423] During the simulation reproduction phase, which is later than the start of the simulation, the reproduced simulated driving behavior corresponding to the perception data is output in the perception coverage area. Based on the first virtual simulated vehicle, the virtual simulated driving behavior of the second virtual simulated vehicle is output in the perception blank area. The second virtual simulated vehicle includes the first virtual simulated vehicle.

[0424] In the simulation prediction phase, which is later than the simulation reproduction phase, the predicted simulation driving behavior of the third virtual simulation vehicle is output on the simulation main road based on the virtual simulation driving behavior and the reproduced simulation driving behavior.

[0425] It should be understood that the computer device 1000 described in the embodiments of this application can perform the data processing methods or apparatus described in the preceding embodiments, and will not be repeated here. In addition, the beneficial effects of using the same method will not be repeated here either.

[0426] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the data processing methods or apparatus described in the preceding embodiments, which will not be repeated here. Furthermore, the beneficial effects of using the same methods will also not be repeated.

[0427] The aforementioned computer-readable storage medium can be an internal storage unit of the data processing apparatus or computer device provided in any of the foregoing embodiments, such as a hard disk or memory of the computer device. The computer-readable storage medium can also be an external storage device of the computer device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., provided on the computer device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the computer device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0428] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. A processor of a computer device reads the computer program from the computer-readable storage medium and executes the computer program, enabling the computer device to perform the data processing methods or apparatus described in the preceding embodiments, which will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated here.

[0429] The terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the term "comprising," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or modules, but may optionally include steps or modules not listed, or may optionally include other step units inherent to these processes, methods, apparatuses, products, or devices.

[0430] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0431] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A data processing method, characterized in that, include: In a driving simulation system, the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area are determined in the simulated main road. At the start of the simulation, a first virtual simulation vehicle is generated in the perception blank area based on the regional relationship between the perception coverage area and the perception blank area. During the simulation reproduction phase, which is later than the simulation start time, the reproduced simulated driving behavior corresponding to the perceived data is output in the perceived coverage area, and the virtual simulated driving behavior of the second virtual simulated vehicle is output in the perceived blank area based on the first virtual simulated vehicle; the second virtual simulated vehicle includes the first virtual simulated vehicle. In the simulation prediction phase, which is later than the simulation reproduction phase, the predicted simulation driving behavior of a third virtual simulation vehicle is output on the simulation main road based on the virtual simulation driving behavior and the reproduced simulation driving behavior.

2. The method according to claim 1, characterized in that, The step of generating a first virtual simulation vehicle in the perception blank area based on the regional relationship between the perception coverage area and the perception blank area includes: Based on the regional relationship between the sensing coverage area and the sensing blank area, the initial traffic state corresponding to the sensing blank area is determined; Based on the initial traffic conditions, a first virtual simulation vehicle is generated in the perception blank area; The method further includes: From the sensing data, the initial sensing data of the sensing coverage area at the start time of the simulation is obtained; Based on the initial perception data, an initial simulation vehicle is generated in the perception coverage area; the initial simulation vehicle belongs to the simulation vehicle corresponding to the simulation driving behavior.

3. The method according to claim 2, characterized in that, The total number of the sensing coverage areas is P, where P is a positive integer; The step of determining the initial traffic state corresponding to the sensing blank area based on the regional relationship between the sensing coverage area and the sensing blank area includes: If the regional relationship between the P sensing coverage areas and the sensing blank area is an upstream regional relationship or a midstream regional relationship, then among the P sensing coverage areas, the sensing coverage area that is connected to the sensing blank area and located downstream of the sensing blank area is obtained as the first sensing coverage area; the upstream regional relationship is used to characterize that the sensing blank area is an upstream region of the P sensing coverage areas; the midstream regional relationship is used to characterize that the sensing blank area is located between two sensing coverage areas among the P sensing coverage areas. From the initial sensing data, the first initial sensing data corresponding to the first sensing coverage area is obtained; If the first initial sensing data satisfies the state setting conditions for the sensing blank area, then the initial traffic state corresponding to the sensing blank area is determined based on the first initial sensing data. If the first initial sensing data does not meet the state setting conditions, then the first historical data of the sensing blank area is obtained, and the initial traffic state is determined based on the first historical data.

4. The method according to claim 3, characterized in that, The first sensing coverage area includes A simulated lanes; A is a positive integer; the A simulated lanes include simulated lane E. f f is a positive integer and f is less than or equal to A; The method further includes: In the first initial perception data, the location of the simulated lane E is obtained. f The simulated vehicle G in the simulation f The simulated vehicle G is described. f It belongs to the initial simulation vehicle; If the simulated vehicle G is reproduced f If the total number is at least two, then the first initial sensing data is determined to satisfy the state setting condition; If the simulated vehicle G is reproduced f If the total number is less than or equal to 1, then the simulated lane E is determined. f If the A simulated lanes belong to the simulated blank lanes, then it is determined that the first initial perception data does not meet the state setting conditions.

5. The method according to claim 4, characterized in that, The step of determining the initial traffic state corresponding to the sensing blank area based on the first initial sensing data includes: Identify at least two replica simulation vehicles G f The distance between any two adjacent simulated vehicles in the simulation. Based on the distance between every two adjacent simulated vehicles, the at least two simulated vehicles G are determined. f The average vehicle spacing; The at least two replicated simulation vehicles G f The average vehicle spacing is determined as the simulated lane E. f The corresponding average vehicle spacing; The vehicle density corresponding to the perception blank area is determined based on the average vehicle spacing corresponding to the A simulated lanes. Based on the vehicle density, the initial traffic state corresponding to the sensing blank area is determined in the first basic traffic map corresponding to the first sensing coverage area.

6. The method according to claim 3, characterized in that, Determining the initial traffic state based on the first historical data includes: If the first historical data is not an empty set, then the first starting historical data corresponding to the simulation start time is obtained from the first historical data, and the first starting historical data is determined as the starting traffic state. If the first historical data is an empty set, the initial traffic state is determined based on the target traffic state in the second basic traffic map corresponding to the perceived blank area.

7. The method according to claim 2, characterized in that, The total number of the sensing coverage areas is M, where M is a positive integer; The step of determining the initial traffic state corresponding to the sensing blank area based on the regional relationship between the sensing coverage area and the sensing blank area includes: If the regional relationship between the M sensing coverage areas and the sensing blank area is a downstream regional relationship, then the second historical data of the sensing blank area is obtained; the downstream regional relationship is used to characterize that the sensing blank area is a downstream region of the M sensing coverage areas; If the second historical data is not an empty set, then the second starting historical data corresponding to the simulation start time is obtained from the second historical data, and the second starting historical data is determined as the starting traffic state corresponding to the perception blank area; If the second historical data is an empty set, then among the M sensing coverage areas, the sensing coverage area that is adjacent to the sensing blank area and located upstream of the sensing blank area is obtained as the upstream sensing coverage area, and the starting traffic state is determined based on the upstream sensing coverage area.

8. The method according to claim 7, characterized in that, Determining the initial traffic state based on the upstream sensing coverage area includes: From the initial sensing data, obtain the second initial sensing data corresponding to the upstream sensing coverage area; If the second initial sensing data meets the state setting conditions for the sensing blank area, then the initial traffic state is determined based on the second initial sensing data. If the second initial sensing data does not meet the state setting conditions, the initial traffic state is determined based on the target traffic state in the third basic traffic map corresponding to the sensing blank area.

9. The method according to claim 2, characterized in that, The step of generating a first virtual simulation vehicle in the perception blank area based on the initial traffic state includes: Based on the initial traffic conditions, determine the average vehicle spacing corresponding to the sensing blank area; If the regional relationship is an upstream regional relationship or a midstream regional relationship, then a first virtual simulated vehicle is generated in the perceptual blank area based on the average vehicle spacing corresponding to the perceptual blank area, the downstream edge of the perceptual blank area, and the direction opposite to the driving direction of the simulated main road; the upstream regional relationship is used to characterize that the perceptual blank area is an upstream area of ​​the perceptual coverage area; the midstream regional relationship is used to characterize that the total number of perceptual coverage areas is at least two, and the perceptual blank area is located between two perceptual coverage areas in at least two perceptual coverage areas; If the regional relationship is a downstream regional relationship, then a first virtual simulation vehicle is generated in the perceptual blank area based on the average vehicle spacing corresponding to the perceptual blank area, the upstream edge of the perceptual blank area, and the driving direction; the downstream regional relationship is used to characterize that the perceptual blank area is a downstream area of ​​the perceptual coverage area.

10. The method according to claim 1, characterized in that, The regional relationship is an upstream regional relationship; the upstream regional relationship is used to characterize that the sensing blank area is the upstream region of the sensing coverage area; The step of outputting the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the first virtual simulation vehicle includes: Based on the upstream edge of the perceived blank area and the vehicle generation line within the perceived blank area, a first vehicle generation sub-region is determined within the perceived blank area; the vehicle generation line is perpendicular to the travel direction of the simulated main road. In the sensing data, downstream sensing data corresponding to the second sensing coverage area is obtained; the second sensing coverage area refers to the area in the sensing coverage area that is connected to the sensing blank area and located downstream of the sensing blank area. Based on the downstream sensing data and the first virtual simulation vehicle, a fourth virtual simulation vehicle is generated in the first vehicle generation sub-region. The first virtual simulation vehicle and the fourth virtual simulation vehicle are identified as the second virtual simulation vehicle; In the perceived blank area, a first vehicle removal line perpendicular to the driving direction is determined; the first vehicle removal line is located in the downstream area of ​​the first vehicle generation sub-area, and the distance between the first vehicle removal line and the downstream edge of the perceived blank area is less than the distance between the first starting vehicle and the downstream edge of the perceived blank area; the first starting vehicle refers to the virtual simulation vehicle that is closest to the downstream edge of the perceived blank area among the first virtual simulation vehicles. Based on the autonomous driving model corresponding to the perceived blank area and the first vehicle removal line, the virtual simulation driving behavior of the second virtual simulation vehicle is output in the perceived blank area; the first vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle in the second virtual simulation vehicle that drives to the first vehicle removal line.

11. The method according to claim 10, characterized in that, The method further includes: Obtain the initial autonomous driving model corresponding to the perception blank area, and obtain the third historical data of the perception blank area; If the third historical data is not an empty set, then the parameters in the initial autonomous driving model corresponding to the perception blank area are adjusted according to the third historical data to obtain the autonomous driving model corresponding to the perception blank area. If the third historical data is an empty set, then the parameters in the initial autonomous driving model corresponding to the perception blank area are adjusted according to the road type corresponding to the perception blank area to obtain the autonomous driving model corresponding to the perception blank area.

12. The method according to claim 10, characterized in that, The method further includes: The virtual simulation vehicle that is closest to the downstream edge of the perception blank area in the second virtual simulation vehicle is determined as the downstream vehicle of the perception blank area; Based on the downstream sensing data, the maximum vehicle speed of the downstream vehicle is determined; The virtual simulation vehicles in the second virtual simulation vehicle, excluding the downstream vehicles, are identified as the first upstream vehicles; The maximum vehicle speed of the first upstream vehicle is determined based on the road type corresponding to the perceived blank area. The step of outputting the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the autonomous driving model corresponding to the perception blank area and the first vehicle removal line includes: Based on the autonomous driving model corresponding to the perception blank area, the first vehicle removal line, the maximum vehicle speed of the first upstream vehicle, and the maximum vehicle speed of the downstream vehicle, the virtual simulation driving behavior of the second virtual simulation vehicle is output in the perception blank area.

13. The method according to claim 12, characterized in that, Determining the maximum vehicle speed of the downstream vehicle based on the downstream sensing data includes: If the downstream sensing data indicates that there is a replica simulation vehicle within the second sensing coverage area, then the replica simulation vehicle closest to the upstream edge of the second sensing coverage area among the replica simulation vehicles within the second sensing coverage area is determined as the second upstream vehicle. The speed of the second upstream vehicle is determined as the maximum speed of the downstream vehicle; the simulated lane to which the second upstream vehicle belongs is the same as the simulated lane to which the downstream vehicle belongs. If the downstream sensing data indicates that there is no replicated simulated vehicle within the second sensing coverage area, and the third historical data of the sensing blank area is not an empty set, then the historical vehicle speed is obtained from the third historical data, and the historical vehicle speed is determined as the maximum vehicle speed of the downstream vehicle. If the downstream sensing data indicates that there is no simulated vehicle within the second sensing coverage area, and the third historical data of the sensing blank area is an empty set, then the maximum vehicle speed of the downstream vehicle is determined according to the road type corresponding to the sensing blank area.

14. The method according to claim 1, characterized in that, The regional relationship is a midstream regional relationship; the midstream regional relationship is used to characterize that the total number of the sensing coverage areas is at least two, and the sensing blank area is located between two sensing coverage areas in the at least two sensing coverage areas; The step of outputting the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the first virtual simulation vehicle includes: The simulated vehicle that drives to the perceived blank area from the simulated vehicle within the third perception coverage area is identified as the fifth virtual simulated vehicle; the simulated vehicle within the third perception coverage area is generated by the perception data; the third perception coverage area refers to the area in the at least two perception coverage areas that is connected to the perceived blank area and located downstream of the perceived blank area. The first virtual simulation vehicle and the fifth virtual simulation vehicle are designated as the second virtual simulation vehicle; In the perceived blank area, a second vehicle removal line is determined perpendicular to the travel direction of the simulated main road; the distance between the second vehicle removal line and the downstream edge of the perceived blank area is less than the distance between the second starting vehicle and the downstream edge of the perceived blank area; the second starting vehicle refers to the virtual simulated vehicle among the first virtual simulated vehicles that is closest to the downstream edge of the perceived blank area; Based on the autonomous driving model corresponding to the perceived blank area and the second vehicle removal line, the virtual simulation driving behavior of the second virtual simulation vehicle is output in the perceived blank area; the second vehicle removal line is used to instruct the driving simulation system to delete the virtual simulation vehicle that drives to the second vehicle removal line.

15. The method according to claim 1, characterized in that, The regional relationship is a downstream regional relationship; the downstream regional relationship is used to characterize that the sensing blank area is a downstream region of the sensing coverage area; The step of outputting the virtual simulation driving behavior of the second virtual simulation vehicle in the perception blank area based on the first virtual simulation vehicle includes: The simulated vehicle that drives to the sensing blank area from the simulated vehicle within the upstream sensing coverage area is identified as the sixth virtual simulated vehicle. The simulated vehicle within the upstream sensing coverage area is generated using the sensing data; the upstream sensing coverage area refers to the area within the sensing coverage area that is adjacent to the sensing blank area and located upstream of the sensing blank area. The first virtual simulation vehicle and the sixth virtual simulation vehicle are designated as the second virtual simulation vehicle; Based on the autonomous driving model corresponding to the perception blank area and the downstream edge of the perception blank area, the virtual simulation driving behavior of the second virtual simulation vehicle is output in the perception blank area; the downstream edge of the perception blank area is used to instruct the driving simulation system to delete the virtual simulation vehicle in the second virtual simulation vehicle that drives to the downstream edge of the perception blank area.

16. The method according to claim 1, characterized in that, The step of outputting the predicted simulated driving behavior of the third virtual simulated vehicle on the simulated main road based on the virtual simulated driving behavior and the reproduced simulated driving behavior includes: The virtual simulation vehicle corresponding to the simulated driving behavior that meets the simulation prediction conditions in the virtual simulated driving behavior is identified as the seventh virtual simulation vehicle; the seventh virtual simulation vehicle belongs to the second virtual simulation vehicle; The simulated driving behavior that meets the simulation prediction conditions in the simulated driving behavior is identified as the target simulated driving vehicle; the target simulated driving vehicle is generated based on the perception data. Based on the upstream edge of the simulated main road, a second vehicle generation sub-region is determined in the simulated main road; Based on the fourth historical data of the perception area to which the second vehicle generation sub-region belongs, an eighth virtual simulation vehicle is generated in the second vehicle generation sub-region; the perception area to which the second vehicle generation sub-region belongs belongs to the perception coverage area or the perception blank area. The seventh virtual simulation vehicle, the target reproduction simulation vehicle, and the eighth virtual simulation vehicle are identified as the third virtual simulation vehicle. Based on the autonomous driving model corresponding to the simulated main road and the downstream edge of the simulated main road, the predicted simulated driving behavior of the third virtual simulated vehicle is output on the simulated main road; the downstream edge of the simulated main road is used to instruct the driving simulation system to delete the virtual simulated vehicle in the third virtual simulated vehicle that drives to the downstream edge of the simulated main road.

17. A data processing apparatus, characterized in that, include: The first determining module is used in the driving simulation system to determine the sensing coverage area with sensing data and the sensing blank area that does not overlap with the sensing coverage area in the simulated main road. The first generation module is used to generate a first virtual simulation vehicle in the perception blank area at the start of the simulation, based on the regional relationship between the perception coverage area and the perception blank area. The first output module is configured to output the reproduced simulated driving behavior corresponding to the perceived data in the perceived coverage area during the simulation reproduction phase, which is later than the simulation start time; and to output the virtual simulated driving behavior of the second virtual simulated vehicle in the perceived blank area based on the first virtual simulated vehicle; the second virtual simulated vehicle includes the first virtual simulated vehicle. The second output module is used to output the predicted simulated driving behavior of a third virtual simulated vehicle on the simulation main road, based on the virtual simulated driving behavior and the reproduced simulated driving behavior, in the simulation prediction stage, which is later than the simulation reproduction stage.

18. A computer device, characterized in that, include: Processor, memory, and network interface; The processor is connected to the memory and the network interface, wherein the network interface is used to provide data communication functions, the memory is used to store computer programs, and the processor is used to invoke the computer programs to cause the computer device to perform the method according to any one of claims 1 to 16.

19. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded and executed by a processor to cause a computer device having the processor to perform the method of any one of claims 1-16.

20. A computer program product, characterized in that, The computer program product includes a computer program stored in a computer-readable storage medium, the computer program being adapted to be read and executed by a processor to cause a computer device having the processor to perform the method of any one of claims 1-16.

Images