A method and system for simulating road dynamic scenarios

Abstract

The present invention relates to the field of autonomous driving technology, and provides a method and system for simulating road dynamic scenarios. The method includes: generating a road network by a multi-scenario road network construction module; generating traffic flow on the road network by a multi-source traffic flow generation module, wherein the movement of vehicles is controlled according to the traffic flow; determining the joint behavior of multiple vehicles by a multi-vehicle joint decision-making and planning module, and planning the trajectories of multiple vehicles; and recording simulation information and analyzing the target scenario by a multi-dimensional scenario analysis module. The present invention can be used for long-term continuous simulation under an urban-level road network. For multi-vehicle interaction within a scenario, it can simulate an efficient and detailed dynamic scenario, and is well applicable to autonomous driving ability testing and data generation of road scenarios.

Description

Technical Field

[0001] The present invention generally relates to the field of autonomous driving technology. Specifically, the present invention relates to a method and system for simulating road dynamic scenarios. Background Art

[0002] Road dynamic scenarios include static road environments and dynamically moving vehicles. Simulating road dynamic scenarios requires depicting the parsing and application of road information by vehicles and modeling the multi-vehicle movement interaction in the scenario, which is very crucial for applications such as autonomous driving testing and digital twins of road scenarios. According to a report by the RAND Corporation, it is necessary to conduct at least 11 billion miles of road tests to prove that autonomous vehicles are safer than humans. However, real vehicle road tests have problems such as high costs and difficult-to-guarantee safety. Therefore, simulation testing is an important alternative means. However, existing road simulation tests are usually more targeted at functional tests, that is, testing the performance of algorithms in specific scenarios, and there are problems such as a small test road range, single and fragmented test scenarios, short test time, and lack of attention to the overall performance of the ego vehicle and other vehicles in the scenario.

[0003] Specifically, traffic simulation software represented by SUMO and PTV Vissim can achieve large-scale traffic simulation to display the overall traffic situation of the road network, but the simulation of vehicle behavior is relatively simple. Usually, following and lane-changing models are used to control vehicles, without strict vehicle kinematic constraints, unable to truly reflect the microscopic movement process of vehicles and lacking authenticity. Autonomous driving simulation software represented by Baidu's Apollo and 51World's 51Sm-One has more strict kinematic constraints or dynamic constraints on vehicle trajectory planning, can provide a more realistic driving environment, and can provide a more reasonable test environment for the decision-making and planning algorithms of autonomous driving. However, such simulation software lacks interactive modeling of vehicles and also lacks authenticity. Simulation systems represented by SimNet and InterSim, which combine data-driven and learning algorithms, can implicitly learn vehicle coupling interactions from real collected data sets and simulate the interactive movement of vehicles in a small range. However, such simulations are highly dependent on real collected data, difficult to perform scenario editing and data augmentation, and have the problem of fragmented scenarios. Summary of the Invention

[0004] To at least partially solve the above problems in the prior art, the present invention proposes a method for simulating road dynamic scenarios, including the following steps:

[0005] Generating a road network by a multi-scenario road network construction module;

[0006] Generating a traffic flow on the road network by a multi-source traffic flow generation module, wherein the movement of vehicles is controlled according to the traffic flow;

[0007] The joint behavior of multiple vehicles is determined by the multi-vehicle joint decision-making and planning module, and the trajectories of multiple vehicles are planned; and

[0008] The multi-dimensional scenario analysis module records simulation information and analyzes the target scenario.

[0009] In an embodiment of the present invention, it is stipulated that the road network generated by the multi-scenario road network construction module includes:

[0010] The road network construction unit generates the topological information of the road network;

[0011] The lane-level path construction unit generates lane-level paths; and

[0012] The scene range construction unit generates a scene range, wherein the movement of vehicles is controlled according to the traffic flow outside the scene range, and refined trajectory planning is performed on the vehicles within the scene range.

[0013] In an embodiment of the present invention, it is stipulated that the traffic flow generated by the multi-source traffic flow generation module on the road network includes:

[0014] The baseline model generation unit generates traffic flow according to the baseline model, wherein the baseline model includes a car-following model, a lane-changing model, and an intersection passing model;

[0015] The road-sourced data generation unit generates traffic flow according to road-sourced data, wherein the road-sourced data includes data collected by roadside fixed sensors, data collected by in-vehicle sensors, and data collected by unmanned aerial vehicle aerial photography; and

[0016] The custom scenario unit custom-generates traffic flow according to road scene standards.

[0017] In an embodiment of the present invention, it is stipulated that the multi-vehicle joint decision-making and planning module determines the joint behavior of multiple vehicles and plans the trajectories of multiple vehicles, including:

[0018] The trajectory prediction unit predicts the trajectory of a vehicle at a future moment based on the trajectories of the vehicle at historical moments and the current moment;

[0019] The multi-vehicle joint behavior decision-making unit performs traffic flow grouping, in-group joint behavior decision-making, and decision behavior benefit evaluation on multiple vehicles; and

[0020] The trajectory planning unit performs parallel trajectory planning on multiple vehicles, wherein the trajectory planning includes target point sampling, optional trajectory generation, trajectory benefit evaluation, and optimal trajectory generation; and / or

[0021] The multi-dimensional scenario analysis module records simulation information and analyzes the target scenario, including:

[0022] The simulation information is recorded by a simulation recording unit, and the simulation information includes road network information and vehicle information;

[0023] The vehicle state is analyzed by a vehicle state analysis unit; and

[0024] A target scene is analyzed by a scene analysis unit.

[0025] In an embodiment of the present invention, it is stipulated that the topological information of the road network generated by a road network construction unit includes:

[0026] A discrete point set is generated according to the acquisition information of a sensor, and the discrete point set;

[0027] Road markings are generated according to the discrete point set, and the road markings include lane dividing lines, road boundary lines, stop lines, and turning signs; and

[0028] The upstream and downstream connection relationships of roads and road convergence areas are determined to generate the topological information of the road network; and / or

[0029] The lane-level path generated by the lane-level path construction unit includes:

[0030] A path is determined on the road network according to the start and end point information of the vehicle;

[0031] The path is split into a plurality of interconnected path components, and the path components include road sections or road sections and their downstream intersections, and a plurality of lanes are included on the road sections; and

[0032] Feasible lanes are determined on the path components; and / or

[0033] The scene range includes:

[0034] A first scene range, which includes the area within a circle centered on the target vehicle and with a specific distance as the radius;

[0035] A second scene range, which includes the path component where the target vehicle is located;

[0036] A third scene range, which includes the area surrounded by a plurality of surrounding vehicles closest to the target vehicle.

[0037] In an embodiment of the present invention, it is stipulated that the lane-level path construction unit determines feasible lanes on the path components, including:

[0038] In the first stage, all lanes on the path component are determined as feasible lanes, and the condition for the vehicle to be in the first stage is expressed by the following formula:

[0039]

[0040] Where D look_forwardRepresents the vehicle's perception range of the front, S remain Represents the remaining length of the road segment, L lane Represents the total length of the road segment;

[0041] In the second stage, determine the feasible lanes on the path component according to the turning information of the path, where the condition for the vehicle to be in the second stage is expressed as the following formula:

[0042]

[0043] In the third stage, make the vehicle drive along the lane it is in, where the condition for the vehicle to be in the third stage is expressed as the following formula:

[0044] S remain <D no_change

[0045] where D no_change Represents the shortest distance required for the vehicle to change lanes; and

[0046] In the fourth stage, make the vehicle drive along the lane it is in, where the vehicle is in the fourth stage when it is inside the intersection.

[0047] In an embodiment of the present invention, it is stipulated that grouping multiple vehicles into traffic flows includes:

[0048] Determine whether there is a potential conflict between any two adjacent vehicles in the continuous traffic flow according to the current speed, relative distance, maximum acceleration and deceleration, and minimum safety distance of the vehicles;

[0049] When there is a potential conflict between two vehicles, determine that the two vehicles will interact; and

[0050] Group multiple vehicles in the continuous traffic flow into traffic flows through the above interaction determination so that any two vehicles within the group have direct interaction or indirect interaction; and / or

[0051] Performing joint decision-making within the group includes:

[0052] Using a Monte Carlo search tree, combined with vehicle positions and high-precision road network information, determine the optional behaviors of each vehicle, and then generate meta-nodes including non-conflicting joint behaviors of multiple vehicles; and

[0053] Generate a multi-vehicle joint behavior decision tree composed of multiple meta-nodes according to multiple decision time steps; and / or

[0054] Performing decision behavior benefit evaluation includes:

[0055] Make each vehicle evaluate the benefit of its own behavior according to the degree of completion of its own decision-making goal and the safety, comfort, and efficiency indicators associated with the behavior;

[0056] Have each vehicle weigh the benefits of its own vehicle and other vehicles to obtain a weighted benefit; and

[0057] Determine the sum of the weighted benefits of all vehicles in the group to obtain a grouped benefit, and determine the grouped multi-vehicle joint behavior according to the grouped benefit.

[0058] The present invention also provides a road dynamic scenario simulation system, including:

[0059] A multi-scenario road network construction module configured to generate a road network;

[0060] A multi-source traffic flow generation module configured to generate a traffic flow on the road network, wherein the movement of vehicles is controlled according to the traffic flow;

[0061] A multi-vehicle joint decision-making and planning module configured to determine the joint behavior of multiple vehicles and plan the trajectories of multiple vehicles; and

[0062] A multi-dimensional scenario analysis module configured to record simulation information and analyze the target scenario.

[0063] The present invention also provides a computer system, including:

[0064] A processor configured to execute machine-readable instructions; and

[0065] A memory storing machine-readable instructions that, when executed by the processor, perform the steps according to the method.

[0066] The present invention also provides a computer-readable storage medium having stored thereon machine-readable instructions that, when executed by a processor, perform the steps according to the method.

[0067] The present invention has at least the following beneficial effects: The present invention provides a road dynamic scenario simulation method and system, which can be used for long-term continuous simulation under an urban-level road network. For multi-vehicle interaction in a scenario, it can simulate an efficient and fine dynamic scenario, and is well applicable to autonomous driving ability testing and road scenario data generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] To further clarify the advantages and features of the embodiments of the present invention, a more specific description of the embodiments of the present invention will be presented with reference to the accompanying drawings. It can be understood that these drawings only depict typical embodiments of the present invention and will not be considered as limiting its scope. In the drawings, for clarity, the same or corresponding components will be denoted by the same or similar reference numerals.

[0069] Figure 1 Shows a computer system implementing the method and / or system according to the present invention.

[0070] Figure 2 Shows a schematic diagram of a road dynamic scenario simulation system in an embodiment of the present invention.

[0071] Figure 3 Shows a schematic diagram of constructing a high-precision road network according to road geometric information and topological information in an embodiment of the present invention.

[0072] Figure 4 Shows a schematic diagram of a road network of different road types in an embodiment of the present invention.

[0073] Figure 5 Shows a schematic diagram of constructing lane-level path information in an embodiment of the present invention.

[0074] Figure 6 Shows a schematic diagram of available lanes in an embodiment of the present invention.

[0075] Figure 7 Shows a schematic diagram of a scene range in an embodiment of the present invention.

[0076] Figure 8 Shows a flowchart of the operation of a multi-vehicle joint behavior decision-making unit in an embodiment of the present invention.

[0077] Figure 9 Shows a schematic diagram of the structure of a simulation recording unit in an embodiment of the present invention.

[0078] Figure 10 Shows a schematic diagram of a road dynamic scenario simulation method in an embodiment of the present invention. Detailed implementation manners

[0079] It should be noted that the components in the respective drawings may be exaggerated for illustration purposes and are not necessarily to scale correctly. In the respective drawings, the same or functionally identical components are provided with the same reference numerals.

[0080] In the present invention, unless otherwise specified, "arranged on...", "arranged above...", and "arranged over..." do not exclude the presence of intermediate objects therebetween. In addition, "arranged on or above..." only represents the relative positional relationship between two components, and in certain cases, such as after reversing the product direction, it can also be converted to "arranged under or below...", and vice versa.

[0081] In the present invention, the respective embodiments are only intended to illustrate the solutions of the present invention and should not be construed as restrictive.

[0082] In the present invention, unless otherwise specified, the quantifiers "a" and "one" do not exclude the scenario of multiple elements.

[0083] It should also be noted here that in the embodiments of the present invention, for the sake of clarity and simplicity, only a part of the components or assemblies may be shown. However, those of ordinary skill in the art can understand that under the teaching of the present invention, the required components or assemblies can be added according to the specific scenario requirements. In addition, unless otherwise specified, the features in different embodiments of the present invention can be combined with each other. For example, a certain feature in the second embodiment can be used to replace the corresponding or functionally identical or similar feature in the first embodiment, and the obtained embodiment also falls within the disclosure scope or the recorded scope of the present application.

[0084] It should also be noted here that within the scope of the present invention, terms such as "identical", "equal", "equal to" do not mean that the two values are absolutely equal, but allow a certain reasonable error. That is to say, these terms also cover "substantially identical", "substantially equal", "substantially equal to". By analogy, in the present invention, terms indicating directions such as "perpendicular to" and "parallel to" also cover the meanings of "substantially perpendicular to" and "substantially parallel to".

[0085] In addition, the numbering of the steps of the various methods of the present invention does not limit the execution order of the method steps. Unless specifically stated, the method steps can be executed in different orders.

[0086] The present invention will be further described below with reference to the accompanying drawings in conjunction with the specific embodiments.

[0087] Figure 1 A computer system 100 for implementing the method and / or system according to the present invention is shown. Unless otherwise specified, the method and / or system according to the present invention can be executed in the Figure 1 shown computer system 100 to achieve the purpose of the present invention, or the present invention can be distributedly implemented in multiple computer systems 100 according to the present invention through a network, such as a local area network or the Internet. The computer system 100 of the present invention can include various types of computer systems, such as handheld devices, laptop computers, personal digital assistants (PDAs), multi-processor systems, microprocessor-based or programmable consumer electronic devices, network PCs, minicomputers, mainframes, network servers, tablet computers, and so on.

[0088] As Figure 1As shown, computer system 100 includes a processor 111, a system bus 101, a system memory 102, a video adapter 105, an audio adapter 107, a hard drive interface 109, an optical drive interface 113, a network interface 114, and a universal serial bus (USB) interface 112. The system bus 101 can be any of several types of bus structure types, such as a memory bus or memory controller, a peripheral bus, and a local bus using various bus architectures. The system bus 101 is used for communication between various bus devices. In addition to Figure 1 the bus devices or interfaces shown, other bus devices or interfaces are also conceivable. The system memory 102 includes a read-only memory (ROM) 103 and a random access memory (RAM) 104. The ROM 103 can store, for example, basic input / output system (BIOS) data for basic routines for information transfer at startup, while the RAM 104 is used to provide relatively fast operating memory for the system. The computer system 100 also includes a hard drive 109 for reading and writing to the hard drive 110, an optical drive interface 113 for reading and writing to optical media such as CD-ROMs, and so on. The hard drive 110 can store, for example, an operating system and application programs. The drive and its associated computer-readable medium provide non-volatile storage of computer-readable instructions, data structures, program modules, and other data for the computer system 100. The computer system 100 can also include a video adapter 105 for image processing and / or image output, which is used to connect output devices such as a display 106. The computer system 100 can also include an audio adapter 107 for audio processing and / or audio output, which is used to connect output devices such as speakers 108. In addition, the computer system 100 can also include a network interface 114 for network connection, where the network interface 114 can be connected to the Internet 116 through a network device such as a router 115, and the connection can be wired or wireless. Additionally, the computer system 100 can also include a universal serial bus interface (USB) 112 for connecting peripheral devices, where the peripheral devices include, for example, a keyboard 117, a mouse 118, and other peripheral devices such as a microphone, a camera, etc.

[0089] When the present invention is implemented on Figure 1 the computer system 100 described above, it can be used for long-term continuous simulation under an urban-level road network. For multi-vehicle interaction in a scenario, it can simulate an efficient and detailed dynamic scenario and is well-suited for autonomous driving ability testing and data generation for road scenarios.

[0090] In addition, each embodiment may be provided as a computer program product that may include one or more machine-readable media having machine-executable instructions stored thereon. When these instructions are executed by one or more machines, such as a computer, a computer network, or other electronic devices, the one or more machines may be caused to perform the operations in accordance with the embodiments of the present invention. Machine-readable media may include, but are not limited to, floppy disks, optical disks, CD-ROMs (Compact Disc Read-Only Memories), and magneto-optical disks, ROMs (Read-Only Memories), RAMs (Random Access Memories), EPROMs (Erasable Programmable Read-Only Memories), EEPROMs (Electrically Erasable Programmable Read-Only Memories), magnetic or optical cards, flash memories, or other types of media / machine-readable media suitable for storing machine-executable instructions.

[0091] In addition, each embodiment may be downloaded as a computer program product, wherein the program may be transmitted from a remote computer (e.g., a server) to a requesting computer (e.g., a client) by one or more data signals embodied in and / or modulated by a carrier wave or other propagation medium via a communication link (e.g., a modem and / or a network connection). Thus, the machine-readable media used herein may include such a carrier wave, but this is not required.

[0092] In the present invention, the modules of the system according to the present invention may be implemented using software, hardware, firmware, or a combination thereof. When a module is implemented using software, the functions of the module may be implemented through a computer program process. For example, the module may be implemented by a code segment (such as a code segment in languages such as C, C++) stored in a storage device (such as a hard disk, memory, etc.), wherein when the code segment is executed by a processor, the corresponding functions of the module can be implemented. When a module is implemented using hardware, the functions of the module may be implemented by setting the corresponding hardware structure. For example, the functions of the module may be implemented by hardware programming of a programmable device such as a Field Programmable Gate Array (FPGA), or by designing an Application Specific Integrated Circuit (ASIC) including electronic devices such as a plurality of transistors, resistors, and capacitors. When a module is implemented using firmware, the functions of the module may be written in a read-only memory such as an EPROM or EEPROM of the device in the form of program code, and when the program code is executed by a processor, the corresponding functions of the module can be implemented. In addition, certain functions of the module may need to be implemented by separate hardware or in cooperation with the hardware. For example, the detection function is implemented by a corresponding sensor (such as a proximity sensor, an acceleration sensor, a gyroscope, etc.), the signal transmission function is implemented by a corresponding communication device (such as a Bluetooth device, an infrared communication device, a baseband communication device, a Wi-Fi communication device, etc.), the output function is implemented by a corresponding output device (such as a display, a speaker, etc.), and so on.

[0093] Figure 10Shows a schematic diagram of a road dynamic scenario simulation method in an embodiment of the present invention. As Figure 10 shown, the method may include the following steps:

[0094] Step 1001: Generate a road network by the multi-scenario road network construction module 201.

[0095] Step 1002: Generate a traffic flow on the road network by the multi-source traffic flow generation module 202, and control the movement of vehicles according to the traffic flow.

[0096] Step 1003: Determine the joint behavior of multiple vehicles by the multi-vehicle joint decision-making and planning module 203, and plan the trajectories of multiple vehicles.

[0097] Step 1004: Record simulation information by the multi-dimensional scenario analysis module 204, and analyze the target scenario.

[0098] Figure 2 Shows a schematic diagram of a road dynamic scenario simulation system in an embodiment of the present invention. As Figure 2 shown, the road dynamic scenario simulation system 200 may include a multi-scenario road network construction module 201, a multi-source traffic flow generation module 202, a multi-vehicle joint decision-making and planning module 203, and a multi-dimensional scenario analysis module 204.

[0099] The multi-scenario road network construction module 201 includes a road network construction unit 2011, a lane-level path construction unit 2012, and a scenario range construction unit 2013. Among them, the road network construction unit 2011 is configured to restore the geometric information of the road through data such as images and point clouds collected by multi-type sensors, and generate the topological information of the road network according to trajectory data or expert experience. These information can construct a high-precision road network covering multi-type road scenarios such as straight roads, intersections, roundabouts, and ramps. The lane-level path construction unit 2012 is configured to plan a route according to the start and end point information of the vehicle, determine the feasible lanes of each section of the vehicle on the route, and then construct a lane-level path. The scenario range construction unit 2013 is configured to make the scenario range follow the target vehicle to move, so as to continuously simulate the target scenario (the first type of target scenario, the position can change) where the target vehicle is located; or the scenario range construction unit 2013 is configured to fix the scenario range in a specific road scenario in the road network (the second type of target scenario, the position does not change), so as to continuously simulate the dynamic changes of the target road scenario.

[0100] The multi-source traffic flow generation module 202 includes a baseline model generation unit 2021, a road-collected data generation unit 2022, and a custom scenario unit 2023. The baseline model generation unit 2021 is configured to generate traffic flow according to baseline traffic models such as car-following, lane-changing, and intersection passing. The road-collected data generation unit 2022 is configured to generate traffic flow according to real data collected by roadside sensors. The custom scenario unit 2023 is configured to custom-generate traffic flow according to road scenario standards.

[0101] The multi-vehicle joint decision-making and planning module 203 includes a trajectory prediction unit 2031, a multi-vehicle joint behavior decision-making unit 2032, and a trajectory planning unit 2033. The trajectory prediction unit 2031 includes a vehicle trajectory prediction interface and a baseline model. The trajectory prediction unit 2031 is configured to predict vehicle behavior and trajectories in real time during the simulation process. The multi-vehicle joint behavior decision-making unit 2032 is configured to provide a multi-vehicle decision-making combination for behaviors such as maintaining lane, changing lanes left and right, accelerating and decelerating, and overtaking, which can reflect the social interaction and high efficiency of multi-vehicle joint decision-making. The trajectory planning unit 2033 is configured to provide trajectory planning for multiple vehicles in parallel, which can take into account measurement indicators such as safety, efficiency, and comfort.

[0102] The multi-dimensional scenario analysis module 204 includes a simulation recording unit 2041, a vehicle state analysis unit 2042, and a scenario analysis unit 2043. The simulation recording unit 2041 is configured to record in real time road network information, vehicle attributes, vehicle motion parameters, etc. during the entire process of the simulation scenario. The recorded information files can be used for scenario playback and generation of scenario simulation data. The vehicle state analysis unit 2042 is configured to analyze the state of the target vehicle during the simulation process, which can provide indicators such as safety, comfort, energy consumption, and trajectory quality. The scenario analysis unit 2043 is configured to comprehensively evaluate the target scenario during the simulation process online or offline, which can provide indicators such as scenario complexity and danger.

[0103] The following will introduce each module and unit in the embodiments of the present invention in detail with reference to the accompanying drawings.

[0104] The multi-scenario road network construction module 201 includes a road network construction unit 2011, a lane-level path construction unit 2012, and a scenario scope construction unit 2013.

[0105] City-level road network support is the basis for long-term simulation. The road network construction unit 2011 is configured to restore the geometric information of the road through data such as images and point clouds collected by multiple types of sensors, and generate the topological information of the road network according to trajectory data or expert experience. These information can be used to construct a high-precision road network covering multiple types of road scenarios such as straight roads, intersections, roundabouts, and ramps.

[0106] The road network construction unit 2011 can perform the following actions: obtaining a discrete point set related to road markings based on sensors such as lidar; restoring road markings according to the discrete point set through semantic segmentation and curve fitting methods, where the road markings include geometric information such as lane demarcation lines, road boundary lines, stop lines, and turning signs; determining the upstream and downstream connection relationships of roads and road convergence areas according to trajectory data or expert experience, generating topological information of the road network, and further determining semantic map information such as lane centerlines, intersections, and ramps.

[0107] Further, to meet the requirements of decision-making and planning tasks, the road network construction unit 2011 can construct Frenet coordinate systems for each section of the road network and road convergence areas. Among them, on the road section, a Frenet coordinate system can be constructed based on the lane lines and their vertical directions. In the road convergence area, lanes within the road convergence area can be constructed based on the end point of the upstream lane and the start point of the downstream lane in the road convergence area. Figure 3 FIG. shows a schematic diagram of constructing a high-precision road network according to road geometric information and topological information in an embodiment of the present invention. As Figure 3 shown, where 301 and 302 are the start and end points of lane L 1 respectively, 303 and 304 are the start and end points of lane L 2 respectively. Lane lines 305 and 306 can be obtained through curve fitting, and then Frenet coordinate systems based on the lanes can be constructed respectively. Since L 1 and L 2 are two connected lanes, the downstream lane of L 1 is L 2 . Selecting the end point 302 of L 1 and the start point 303 of L 2 , lane line 307 within the road convergence area can be obtained through curve fitting, and then a Frenet coordinate system for the lane within the road convergence area can be constructed.

[0108] Figure 4 FIG. shows a schematic diagram of road networks of different road types in an embodiment of the present invention. The above method of constructing Frenet coordinate systems in road sections and road convergence areas can be extended to various road types such as intersections 401, roundabouts 402, and ramps 403, etc., and thus a city-level high-precision road network can be obtained to provide road network support for long-term simulation.

[0109] When performing simulation on a city-level road network, it is necessary to specify a driving path for the vehicle. The lane-level path construction unit 2012 can perform route planning according to the start and end point information of the vehicle, determine the feasible lanes for each section of the vehicle on the route, and further construct a lane-level path.

[0110] Figure 5Shows a schematic diagram of constructing lane-level path information in an embodiment of the present invention. As Figure 5 shown, when simulating on an urban road network, usually the information of the road sections passed by the vehicle is defined, and the path information is determined on the road network 501 through the starting point 502 and the ending point 503 of the vehicle. In order to enable the vehicle to drive correctly on each road section, it is also necessary to determine the lane-level path information, thereby constructing a feasible lane area 504. When the vehicle approaches an intersection, it is necessary to determine a reasonable lane according to the turning information of the planned path at the intersection, so as to complete lane change in advance.

[0111] In the embodiment of the present invention, the path guidance of the vehicle can be realized by providing feasible lanes. Among them, a long-distance path can be split into multiple interconnected path components, and the path components can include a single road section, or include a single road section and its downstream intersection. On the current path component, the vehicle starts driving from the starting point of the road section, enters the downstream intersection, and then leaves the downstream intersection and enters the next path component.

[0112] Figure 6 Shows a schematic diagram of a feasible lane in an embodiment of the present invention. As Figure 6 shown, when determining the feasible lane area on the path component, the driving of the vehicle on the road section can be divided into 4 stages. In the first stage 601, the feasible lane area covers all lanes on the road section, and the condition for the vehicle to be in the first stage 601 can be expressed as the following formula:

[0113]

[0114] Among them, D look_forward represents the vehicle's perception range of the front. S remain represents the remaining length of the road section, that is, the length from the current position of the vehicle to the end of the road section. L lane represents the total length of the road section. That is to say, in the first stage 601, when the vehicle just enters an ordinary lane, if the remaining length of the road section is long, the vehicle can freely choose a lane, and all lanes on the current road section can be used as target lanes.

[0115] When the vehicle has traveled on the road section for a period of time and the remaining length of the road section shortens, and the vehicle needs to enter the target lane that meets the turning requirements of the planned path, otherwise it will not be able to reach the next path component, the vehicle is in the second stage 602. In the second stage 602, the feasible area is the lane on the road section determined by the turning information of the current path, and the condition for the vehicle to be in the second stage 602 can be expressed as the following formula:

[0116]

[0117] When the vehicle approaches the downstream intersection and is unable to change lanes and can only drive along the current lane, the vehicle is in the third stage 603. In the third stage 603, the feasible region is the lane within the road convergence region determined according to the path turning information. The condition for the vehicle to be in the third stage 603 can be expressed as the following formula:

[0118] S remain <D no_change

[0119] where D no_change represents the shortest distance required for the vehicle to change lanes. That is, when the remaining lane length is less than D no_change the vehicle cannot change lanes.

[0120] When the vehicle is inside the intersection, that is, driving on the lane within the road convergence region, the vehicle is in the fourth stage 604. In the fourth stage 604, to ensure the order within the intersection, the feasible vehicle region of the vehicle within the intersection is the current lane, and the vehicle is not allowed to change lanes; and to ensure the continuity of the trajectory planning and let the vehicle smoothly enter the next path component, the feasible lane region should also consider all the lanes within the section of the next path component.

[0121] The scenario range construction unit 2013 is configured to provide a scenario range, which is used to define the range that needs to finely describe the road scenario. Outside the scenario range, the movement of all vehicles is controlled by the traffic flow generation module. Within the scenario range, fine trajectory planning can be performed for all vehicles.

[0122] Figure 7 shows a schematic diagram of a scenario range in an embodiment of the present invention. As Figure 7 shown, the scenario range may include a first scenario range 701, a second scenario range 702, and a third scenario range 703. The first scenario range 701 is the area within a circle centered on the target vehicle with a specific distance as the radius. The second scenario range 702 is the path component where the target vehicle is located. The third scenario range 703 is the area enclosed by no more than a specific number of surrounding vehicles closest to the target vehicle.

[0123] The multi-source traffic flow generation module 202 includes a baseline model generation unit 2021, a road sampling data generation unit 2022, and a custom scenario unit 2023.

[0124] The baseline model generation unit 2021 is configured to generate traffic flow according to baseline traffic models such as car-following, lane-changing, and intersection passing. The car-following model may include a stimulus-response model, a full speed difference model, an intelligent driver model, etc. The role of the car-following model is to keep a reasonable safe distance between vehicles on the lane. The lane-changing model may include MOBIL, LC2013 model, etc. The purpose of the lane-changing model is to enable vehicles to change lanes as necessary following the planned path. The intersection passing model includes a signalized intersection passing model and an unsignalized intersection passing model. The purpose of the intersection passing model is to avoid collisions of vehicles in the road convergence area.

[0125] These baseline models can generate vehicles arriving at the road network in advance according to traffic demands and perform path planning. Generating traffic flow according to the baseline model can avoid performing fine trajectory planning for all vehicles in the road network and save computing resources.

[0126] The road-collected data generation unit 2022 can generate traffic flow according to the real data collected by the roadside. The road-collected data includes data collected by roadside fixed sensors (such as NGSIM datasets), data collected by on-vehicle sensors (such as Waymo datasets), data collected by means of drone aerial photography (such as CitySim datasets), etc. Vehicle calibration and trajectory extraction of the road-collected data can obtain the time-series trajectory data of each vehicle, and by combining with the road network, a real road scene can be reproduced.

[0127] The custom scenario unit 2023 can custom-generate traffic flow according to road scenario standards. For example, it can manually define the paths and trajectories of each vehicle in the scenario according to road scenario-related standards such as OpenScenario. The paths and trajectories of the vehicles include all vehicle state information under the complete time series.

[0128] The multi-vehicle joint decision-making and planning module 203 includes a trajectory prediction unit 2031, a multi-vehicle joint behavior decision-making unit 2032, and a trajectory planning unit 2033.

[0129] The trajectory prediction unit 2031 includes a vehicle trajectory prediction interface and a baseline model, which can predict vehicle behavior and trajectories in real time during the simulation process. The trajectory prediction unit 2031 performs trajectory prediction on the vehicles within the scenario range. The input of the trajectory prediction is the trajectories of the vehicle at the historical moment and the current moment, and the output is the trajectory of the vehicle at the future moment. The baseline model for trajectory prediction is a car-following model based on a vehicle queue. The logic of the car-following model includes: all vehicles keep driving in the current lane; when the vehicle is the leading vehicle in the queue, it drives at a constant speed at the current speed; when the vehicle is not the leading vehicle in the queue, it drives according to the car-following model and maintains a reasonable safe distance from the vehicle in front.

[0130] The multi-vehicle joint behavior decision-making unit 2032 can provide multi-vehicle decision-making combinations for behaviors such as maintaining the lane, changing lanes left and right, accelerating and decelerating, and overtaking, which can reflect the social interaction of multi-vehicles and the efficiency of joint decision-making. Figure 8 shows the workflow of a multi-vehicle joint behavior decision-making unit in an embodiment of the present invention. As Figure 8 shown, the multi-vehicle joint behavior decision-making unit 2032 can perform traffic flow grouping, perform intra-group joint decision-making, and perform decision-making behavior benefit evaluation.

[0131] Among them, traffic flow grouping includes: for any two adjacent vehicles in a continuous traffic flow, determining whether there is a potential conflict between them, where factors such as the current speed, relative distance, maximum acceleration and deceleration, and minimum safety distance of the vehicle are used to judge whether the two adjacent vehicles may conflict in a future period of time; when there is a potential conflict between the two vehicles, it is considered that the two vehicles will interact; through pairwise interaction determination, multiple vehicles in the continuous traffic flow can be divided into groups including a finite number of vehicles, so that any two vehicles within the group have direct interaction or indirect interaction. At the same time, the maximum number of vehicles within the group needs to be considered during the traffic flow grouping process, so that the number of vehicles within each group is not too large.

[0132] Intra-group joint decision-making is a multi-step decision-making in a long cycle, and it can use Monte Carlo search trees for decision-making. Combining vehicle positions and high-precision road network information, determine the optional behaviors of each vehicle, and thus generate meta-nodes covering non-conflicting joint behaviors of multiple vehicles. Considering multiple decision time steps, generate a multi-vehicle joint behavior decision tree composed of multiple meta-nodes.

[0133] Decision-making behavior benefit evaluation is used to determine the best decision-making behavior to be selected under intra-group joint decision-making. For each vehicle, on the one hand, the vehicle will evaluate the benefit of its own behavior according to the degree of completion of its own decision-making goal (such as whether a specific lane-changing behavior is completed) and indicators such as safety, comfort, and efficiency associated with the behavior; on the other hand, the social value orientation of the vehicle indicates that the vehicle weighs the benefits of its own vehicle and other vehicles during the decision-making process to obtain a weighted benefit. For all vehicles within the group, it is necessary to determine the sum of the weighted benefits of all vehicles to obtain the group benefit, and thus determine the multi-vehicle joint behavior of the group. For different groups, multi-vehicle joint decision-making planning can be carried out in parallel.

[0134] The trajectory planning unit 2033 can provide trajectory planning for multiple vehicles in parallel, and it can take into account measurement indicators such as safety, efficiency, and comfort. Among them, under the guidance of behavior decision-making, the trajectory planning unit 2033 generates appropriate vehicle driving trajectories, and the trajectory planning includes steps such as target point sampling, optional trajectory generation, trajectory benefit evaluation, and best trajectory determination. The trajectory planning between each vehicle is independent of each other, so it can be carried out in parallel.

[0135] The multi-dimensional scenario analysis module 204 includes a simulation recording unit 2041, a vehicle state analysis unit 2042, and a scenario analysis unit 2043.

[0136] The simulation recording unit 2041 can record information such as road network information, vehicle attributes, and vehicle motion parameters throughout the entire process of the simulation scenario in real time. The file for recording the information can be used for scenario playback and generation of scenario simulation data. Figure 9 The figure shows a schematic diagram of the structure of a simulation recording unit in an embodiment of the present invention. As Figure 9 shown, the information recorded by the simulation recording unit 2041 can include road network information and vehicle information.

[0137] The road network information includes road network geometric structure information and topological structure information. The basic unit of the road network is a lane, and the geometric structure of the lane can be described by the set of centerline points of each lane. The topological structure information includes the subordinate relationship of the lane to the road section or intersection and the connection relationship between the upstream and downstream of the lane.

[0138] The vehicle information includes the fixed attributes and real-time state information of the vehicle. The fixed attributes of the vehicle include the three-dimensional size of the vehicle, maximum speed, maximum acceleration, maximum deceleration, path information of the vehicle, etc. The real-time state information of the vehicle is recorded in chronological order, and it includes the coordinate position, speed, acceleration, heading angle, ID of the lane where the vehicle is located, relative position on the lane where the vehicle is located, and relative position on the path where the vehicle is located at each time step.

[0139] The vehicle state analysis unit 2042 is configured to analyze the state of the target vehicle during the simulation process, and it can provide indicators such as safety, comfort, energy consumption, and trajectory quality.

[0140] The time to collision can be used to analyze the safety of the target vehicle. The time to collision refers to the time required for the vehicle itself to collide with the vehicle in front when both the target vehicle and the vehicle in front maintain their speeds. When the speed of the target vehicle is less than or equal to the speed of the vehicle in front, the time to collision is infinite. When the speed of the target vehicle is greater than the speed of the vehicle in front, the time to collision is calculated as the distance between the front of the target vehicle and the rear of the vehicle in front divided by the difference in speed between the target vehicle and the vehicle in front. Jerk can be used to analyze the comfort of the target vehicle, where jerk is the derivative of the acceleration of the target vehicle with respect to time. The engine output power can be used to analyze the energy consumption of the target vehicle, where the engine output power is the square of the product of the instantaneous speed and instantaneous acceleration of the target vehicle. The heading angle deviation and trajectory deviation can be used to measure the quality of the trajectory of the target vehicle, where the heading angle deviation is the angle between the heading angle of the target vehicle and the tangent direction of the centerline of the lane where the vehicle is located, and the trajectory deviation is the distance between the centroid of the target vehicle and the centerline of the lane where the vehicle is located.

[0141] The described scenario analysis unit 2043 is configured to comprehensively evaluate the target scenario of the simulation process online or offline, and it can provide indicators such as scenario complexity and danger level.

[0142] The number of interacting vehicles and the road congestion level can be used to determine the scenario complexity. Given the scenario range and the information about the lane where the target vehicle is located, the number of interacting vehicles represents the number of vehicles other than the target vehicle within the scenario range, and the road congestion level represents the average speed of all vehicles within the scenario range.

[0143] The scenario danger level can be analyzed using the duration of the high-risk scenario and the time integral of the high-risk scenario. The high-risk scenario threshold refers to a specific value of the remaining collision time, and the high-risk situation refers to the situation where the remaining collision time is lower than the high-risk scenario threshold. Given the scenario range and the high-risk scenario threshold, the duration of the high-risk scenario refers to the accumulated time when the scenario is in a high-risk situation. The high-risk degree is expressed as the difference between the high-risk scenario threshold and the expected collision time (when the high-risk scenario threshold is less than or equal to the expected collision time, it is a non-high-risk situation, and the high-risk degree takes a value of 0). Given the scenario range and the high-risk scenario threshold, the time integral of the high-risk scenario is expressed as the integral of the high-risk degree of this scenario with respect to time.

[0144] Although the embodiments of the present invention have been described above, it should be understood that they are presented as examples only and not as limitations. It is obvious to those skilled in the relevant art that various combinations, variations, and changes can be made to them without departing from the spirit and scope of the present invention. Therefore, the width and scope of the present invention disclosed herein should not be limited by the above-disclosed exemplary embodiments, but should be defined only by the appended claims and their equivalents.

Claims

1. A method for simulating a road dynamic scenario, characterized in that, it includes the following steps: Generate a road network by a multi-scenario road network construction module; Generate traffic flow on the road network by a multi-source traffic flow generation module, wherein the movement of vehicles is controlled according to the traffic flow; Determine the joint behavior of multiple vehicles by a multi-vehicle joint decision-making and planning module, and plan the trajectories of multiple vehicles; and Record simulation information by a multi-dimensional scenario analysis module and analyze the target scenario; wherein generating a road network by a multi-scenario road network construction module includes: Generate topological information of the road network by a road network construction unit; Generate lane-level paths by a lane-level path construction unit; and Generate a scenario range by a scenario range construction unit, wherein the movement of vehicles is controlled according to the traffic flow outside the scenario range, and refined trajectory planning is performed on vehicles within the scenario range; wherein generating topological information of the road network by a road network construction unit includes: Generate a discrete point set according to the acquisition information of sensors, the discrete point set; Generate road markings according to the discrete point set, the road markings include lane demarcation lines, road boundary lines, stop lines, and turning signs; and Determine the upstream and downstream connection relationships of roads and road convergence areas, and generate topological information of the road network; and Generating lane-level paths by the lane-level path construction unit includes: Determine a path on the road network according to the starting point and ending point information of the vehicle; Split the path into multiple interconnected path components, the path components include road segments or road segments and their downstream intersections, and multiple lanes are included on the road segments; and Determine feasible lanes on the path components; and The scenario range includes: A first scenario range, which includes the area within a circle centered on the target vehicle with a specific distance as the radius; A second scenario range, which includes the path component where the target vehicle is located; A third scenario range, which includes the area surrounded by multiple surrounding vehicles closest to the target vehicle; wherein determining feasible lanes on the path components by the lane-level path construction unit includes: Determine all lanes on the path components as feasible lanes in the first stage, wherein the condition for the vehicle to be in the first stage is expressed by the following formula: Among them, represents the perception range of the vehicle ahead, represents the remaining length of the road section, represents the total length of the road section; Determine the feasible lanes on the path components according to the turning information of the path in the second stage, wherein the condition for the vehicle to be in the second stage is expressed by the following formula: ； Make the vehicle drive along the lane where it is located in the third stage, wherein the condition for the vehicle to be in the third stage is expressed by the following formula: wherein represents the shortest distance required for a vehicle to change lanes; and Make the vehicle drive along the lane where it is located in the fourth stage, wherein the vehicle is in the fourth stage when it is inside the intersection.

2. The method for simulating a road dynamic scenario according to claim 1, characterized in that, Generating traffic flow on the road network by the multi-source traffic flow generation module includes: Generate traffic flow by a baseline model generation unit according to a baseline model, wherein the baseline model includes a car-following model, a lane-changing model, and an intersection passing model; Generate traffic flow by a road acquisition data generation unit according to road acquisition data, wherein the road acquisition data includes data collected by roadside fixed sensors, data collected by in-vehicle sensors, and data collected by drone aerial photography; and Generate traffic flow by a custom scenario unit according to the custom of road scenario standards.

3. The road dynamic scenario simulation method according to claim 1, characterized in that, determining the joint behavior of multiple vehicles by the multi-vehicle joint decision-making and planning module, and planning the trajectories of multiple vehicles includes: predicting the future trajectory of a vehicle by the trajectory prediction unit according to the historical and current trajectories of the vehicle; performing traffic flow grouping on multiple vehicles, joint behavior decision-making within the group, and decision-making behavior benefit evaluation by the multi-vehicle joint behavior decision-making unit; and performing parallel trajectory planning on multiple vehicles by the trajectory planning unit, where the trajectory planning includes target point sampling, optional trajectory generation, trajectory benefit evaluation, and optimal trajectory generation; and / or recording simulation information by the multi-dimensional scenario analysis module and analyzing the target scenario includes: recording simulation information by the simulation recording unit, where the simulation information includes road network information and vehicle information; analyzing the vehicle state by the vehicle state analysis unit; and analyzing the target scenario by the scenario analysis unit.

4. The road dynamic scenario simulation method according to claim 3, characterized in that, performing traffic flow grouping on multiple vehicles includes: determining whether there is a potential conflict between any two adjacent vehicles in the continuous traffic flow according to the current speed, relative distance, maximum acceleration and deceleration, and minimum safety distance of the vehicle; when there is a potential conflict between two vehicles, determining that the two vehicles will interact; and grouping multiple vehicles in the continuous traffic flow through the above interaction determination so that any two vehicles within the group have direct interaction or indirect interaction; and / or performing joint decision-making within the group includes: using a Monte Carlo search tree, combining vehicle positions and high-precision road network information, determining the optional behaviors of each vehicle, and then generating meta-nodes including non-conflicting joint behaviors of multiple vehicles; and generating a multi-vehicle joint behavior decision tree composed of multiple meta-nodes according to multiple decision time steps; and / or performing decision-making behavior benefit evaluation includes: enabling each vehicle to evaluate the benefit of its own behavior according to the completion degree of its own decision-making goal and the safety, comfort, and efficiency indicators associated with the behavior; enabling each vehicle to weigh the benefits of its own vehicle and other vehicles to obtain a weighted benefit; and determining the sum of the weighted benefits of all vehicles within the group to obtain the grouping benefit, and determining the grouped multi-vehicle joint behavior according to the grouping benefit.

5. A road dynamic scenario simulation system, characterized in that, including: a multi-scenario road network construction module configured to generate a road network; a multi-source traffic flow generation module configured to generate a traffic flow on the road network, where the movement of vehicles is controlled according to the traffic flow; a multi-vehicle joint decision-making and planning module configured to determine the joint behavior of multiple vehicles and plan the trajectories of multiple vehicles; and a multi-dimensional scenario analysis module configured to record simulation information and analyze the target scenario; where generating a road network by the multi-scenario road network construction module includes: generating the topological information of the road network by the road network construction unit; generating lane-level paths by the lane-level path construction unit; and generating a scenario range by the scenario range construction unit, where outside the scenario range, the movement of vehicles is controlled according to the traffic flow, and within the scenario range, refined trajectory planning is performed on the vehicles. The topological information of the road network generated by the road network construction unit includes: Generating a discrete point set according to the acquisition information of the sensor, the discrete point set; Generating road markings according to the discrete point set, the road markings including lane dividing lines, road boundary lines, stop lines, and turning signs; and Determining the upstream and downstream connection relationships of roads and road convergence areas, and generating the topological information of the road network; and The lane-level path generated by the lane-level path construction unit includes: Determining a path on the road network according to the starting point and ending point information of the vehicle; Splitting the path into multiple interconnected path components, the path components including road segments or road segments and their downstream intersections, and multiple lanes being included on the road segments; and Determining feasible lanes on the path components; and The scene range includes: A first scene range, which includes the area within a circle centered on the target vehicle and with a specific distance as the radius; A second scene range, which includes the path component where the target vehicle is located; A third scene range, which includes the area enclosed by multiple surrounding vehicles closest to the target vehicle; Wherein determining feasible lanes on the path components by the lane-level path construction unit includes: Determining all lanes on the path components as feasible lanes in the first stage, where the condition for the vehicle to be in the first stage is expressed by the following formula: Among them, represents the vehicle's perception range ahead, represents the remaining length of the road section, represents the total length of the road section; Determining the feasible lanes on the path components according to the turning information of the path in the second stage, where the condition for the vehicle to be in the second stage is expressed by the following formula: ； Making the vehicle travel along the lane where it is located in the third stage, where the condition for the vehicle to be in the third stage is expressed by the following formula: wherein represents the shortest distance required for a vehicle to change lanes; and Making the vehicle travel along the lane where it is located in the fourth stage, where the vehicle is in the fourth stage when it is inside the intersection.

6. A computer system, including: A processor configured to execute machine-readable instructions; and A memory storing machine-readable instructions, the machine-readable instructions performing the steps of the method according to any one of claims 1-4 when executed by the processor.

7. A computer-readable storage medium storing machine-readable instructions thereon, the machine-readable instructions performing the steps of the method according to any one of claims 1-4 when executed by the processor.

Images