Intersection construction method and apparatus, and vehicle and computer-readable storage medium

Abstract

Disclosed in the present application are an intersection construction method and apparatus, and a vehicle and a computer-readable storage medium. The method comprises: acquiring initial intersection information of an intersection to be constructed that corresponds to a vehicle; on the basis of collected data of a plurality of sensors of the vehicle, acquiring real-time perception data of the vehicle corresponding to said intersection; on the basis of the real-time perception data, determining intersection road line prediction information corresponding to said intersection; and on the basis of the intersection road line prediction information, optimizing the initial intersection information, so as to obtain target intersection information. In the present application, perception data of a vehicle is used to perform real-time intersection construction on initial intersection information; and compared with using a high-precision map for intersection construction, using real-time perception data of a vehicle can reduce the influence of obstacles and vehicles in an intersection on intersection construction, thereby improving the accuracy of intersection construction, improving the traffic rate and safety of autonomous vehicles at the intersection, and reducing the cost of intersection construction.

Description

Intersection construction method and device, vehicle and computer readable storage medium

[0001] The present application claims priority from the Chinese patent application No. 2024104739016 filed on April 18, 2024, and entitled "Intersection construction method and device, vehicle and computer readable storage medium", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of automobile technology, in particular to an intersection construction method and device, vehicle and computer readable storage medium. BACKGROUND

[0003] In the process of automatic driving, complex traffic situations such as large traffic flow, various types of vehicles, different shapes of intersections, traffic light information, road obstacles such as bridge piers, and vehicles violating traffic rules are encountered. In order to safely drive in a complex traffic situation environment, an automatic driving vehicle needs to construct an intersection, thereby improving the performance of the intersection automatic driving algorithm and improving the path planning and traffic prediction accuracy of the automatic driving vehicle through the construction of the intersection.

[0004] Currently, the related technology uses a high-precision map to construct an intersection. Due to the limitations of geographic data collection and update speed, and the situation that the road condition information on the map cannot be updated in time due to sudden conditions such as road condition mutations, the intersection constructed by using the high-precision map is inaccurate, and even navigation errors may occur. SUMMARY

[0005] To solve or partially solve the problems in the related technology, the present application provides an intersection construction method, device, vehicle and computer readable storage medium, which can improve the accuracy of intersection construction.

[0006] To achieve the above-mentioned purpose, the present application provides an intersection construction method, comprising:

[0007] Obtaining initial intersection information of a to-be-constructed intersection corresponding to a vehicle;

[0008] Based on the collected data of a plurality of sensors of the vehicle, obtaining real-time perception data of the to-be-constructed intersection corresponding to the vehicle;

[0009] Based on the real-time perception data, determining intersection road line prediction information corresponding to the to-be-constructed intersection;

[0010] Based on the intersection road line prediction information, optimizing the initial intersection information to obtain target intersection information.

[0011] In an embodiment, the step of determining the intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data comprises:

[0012] filtering the real-time perception data based on the initial intersection information to obtain target perception data;

[0013] determining the intersection road line prediction information corresponding to the to-be-constructed intersection based on the target perception data.

[0014] In an embodiment, the step of optimizing the initial intersection information based on the intersection road line prediction information to obtain target intersection information comprises:

[0015] optimizing the initial intersection information based on the intersection road line prediction information to obtain optimized initial intersection information;

[0016] adjusting the optimized initial intersection information based on vehicle information, signal light state information, and obstacle information in the target perception data to obtain the target intersection information.

[0017] In an embodiment, the step of optimizing the initial intersection information based on the intersection road line prediction information to obtain optimized initial intersection information comprises:

[0018] determining the entrance and exit positions corresponding to the to-be-constructed intersection based on the intersection road line prediction information;

[0019] correcting the entrance and exit positions in the initial intersection information based on the entrance and exit positions corresponding to the to-be-constructed intersection to obtain the optimized initial intersection information.

[0020] In an embodiment, the step of determining the entrance and exit positions corresponding to the to-be-constructed intersection based on the intersection road line prediction information comprises:

[0021] determining the entrance and exit positions corresponding to the to-be-constructed intersection based on stop lines in the intersection road line prediction information;

[0022] if there is no stop line in the intersection road line prediction information, determining the entrance and exit positions corresponding to the to-be-constructed intersection based on zebra crossings in the intersection road line prediction information;

[0023] if there is no zebra crossing in the intersection road line prediction information, clustering the target perception information, and determining the entrance and exit positions corresponding to the to-be-constructed intersection based on the clustering result.

[0024] In an embodiment, the step of adjusting the optimized initial intersection information based on vehicle information, signal light state information, and obstacle information in the target perception data to obtain the target intersection information comprises:

[0025] adjusting the exit and entrance information in the optimized initial intersection information based on the signal light state information and the obstacle information;

[0026] determining the exit and entrance direction in the optimized initial intersection information based on the route arrow information and the traffic information corresponding to the vehicle information.

[0027] In an embodiment, the step of filtering the real-time perception data based on the initial intersection information to obtain target perception data comprises:

[0028] obtaining the initial intersection size corresponding to the initial intersection information;

[0029] filtering the real-time perception data based on the initial intersection size to obtain the target perception data.

[0030] In an embodiment, after the step of optimizing the initial intersection information based on the intersection road line prediction information to obtain target intersection information, the method further comprises:

[0031] determining intersection obstacle information and abnormal vehicle information corresponding to the target intersection information based on the real-time perception data;

[0032] determining the drivable area corresponding to the vehicle based on the target intersection information, the intersection obstacle information and the abnormal vehicle information.

[0033] In an embodiment, after the step of determining the drivable area corresponding to the vehicle based on the target intersection information, the intersection obstacle information and the abnormal vehicle information, the method further comprises:

[0034] optimizing the drivable area based on pedestrian information and signal light state information in the real-time perception data to obtain a target drivable area.

[0035] In an embodiment, before the step of obtaining the initial intersection information of the to-be-constructed intersection corresponding to the vehicle, the method further comprises:

[0036] obtaining design dimensions and traffic light signal information corresponding to the to-be-constructed intersection;

[0037] determining the initial intersection information corresponding to the to-be-constructed intersection based on the design dimensions and the traffic light signal information.

[0038] In an embodiment, the step of obtaining the initial intersection information of the to-be-constructed intersection corresponding to the vehicle comprises:

[0039] Send a intersection obtaining request to a service platform based on a current position and a driving direction of a vehicle, and receive initial intersection information fed back by the service platform, wherein the service platform feeds back the initial intersection information according to the current position and the driving direction of the vehicle in the intersection obtaining request.

[0040] In addition, to achieve the above object, the present application also provides a vehicle, which comprises:

[0041] A first obtaining module, configured to obtain initial intersection information of a to-be-constructed intersection corresponding to the vehicle;

[0042] A second obtaining module, configured to obtain real-time sensing data of the to-be-constructed intersection corresponding to the vehicle based on collected data of a plurality of sensors of the vehicle;

[0043] A determining module, configured to determine intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time sensing data;

[0044] An optimizing module, configured to optimize the initial intersection information based on the intersection road line prediction information, and obtain target intersection information.

[0045] In addition, to achieve the above object, the present application also provides a intersection construction device, which comprises a memory, a processor and a intersection construction program stored in the memory and executable on the processor, and the intersection construction program implements the steps of the intersection construction method when executed by the processor.

[0046] In addition, to achieve the above object, the present application also provides a computer readable storage medium, which stores a intersection construction program, and the intersection construction program implements the steps of the intersection construction method when executed by a processor.

[0047] In addition, to achieve the above object, the present application also provides a computer program product, which comprises a intersection construction program, and the intersection construction program implements the steps of the intersection construction method when executed by a processor.

[0048] The application obtains initial intersection information of a to-be-constructed intersection corresponding to a vehicle; then, based on collected data of multiple sensors of the vehicle, real-time perception data of the to-be-constructed intersection corresponding to the vehicle is obtained; then, based on the real-time perception data, intersection road line prediction information corresponding to the to-be-constructed intersection is determined; and then, based on the intersection road line prediction information, the initial intersection information is optimized to obtain target intersection information. The application can perform real-time intersection construction on the initial intersection information through the perception data of the vehicle, and compared with the intersection construction using a high-precision map, the influence of obstacles and vehicles in the intersection on the intersection construction can be reduced through the real-time perception data of the vehicle, the accuracy of the intersection construction is improved, the passing rate and safety of the autonomous vehicle at the intersection are improved, and the cost of the intersection construction is reduced.

[0049] It should be understood that the general description above and the following detailed description are only exemplary and explanatory, and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0050] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the figures, and in which:

[0051] FIG. 1 is a structural schematic diagram of an intersection construction device in a hardware running environment related to an embodiment of the application;

[0052] FIG. 2 is a flow schematic diagram of a first embodiment of an intersection construction method of the application;

[0053] FIG. 3 is a functional module schematic diagram of an embodiment of a vehicle of the application. DETAILED DESCRIPTION

[0054] Embodiments of the application will be described in more detail by referring to the attached drawings. Although embodiments of the application are shown in the drawings, it should be understood that the application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the application more thorough and complete, and to fully convey the scope of the application to those skilled in the art.

[0055] The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the application. The singular forms "a", "an" and "the" used in the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein means and includes any or all possible combinations of one or more associated listed items.

[0056] It should be understood that although the terms "first", "second", "third" and the like can be used herein to describe various information, the information should not be limited to these terms. These terms are only used to distinguish one type of information from another type of information. For example, without departing from the scope of the present application, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly specified.

[0057] Currently, in the process of automatic driving, complex traffic situations such as heavy traffic, various types of vehicles, various shapes of intersections, traffic light information, road obstacles such as bridge piers, and vehicles violating traffic rules will be encountered. In order to safely drive in complex traffic situations, the automatic driving vehicle needs to construct the intersection, thereby improving the performance of the intersection automatic driving algorithm, and improving the path planning and traffic prediction accuracy of the automatic driving vehicle through the construction of the intersection.

[0058] Among them, by modeling the scene of the intersection, the automatic driving vehicle can more accurately plan the path, which helps to improve the path planning accuracy of the automatic driving vehicle. By identifying and analyzing the intersection scene, the automatic driving vehicle can predict the actions of other vehicles, pedestrians, traffic lights and other traffic elements, and improve the traffic flow of the automatic driving vehicle in the intersection scene. By modeling the scene of the intersection, the automatic driving vehicle can obtain more accurate road information, such as lane lines, obstacles, zebra crossings, etc., so as to better give accurate driving instructions. By analyzing the intersection scene, the automatic driving vehicle can understand the possible driving methods under the current traffic situation, and give reasonable driving decisions according to the traffic rules and safety principles, thereby improving the decision-making ability of the automatic driving vehicle. By modeling and analyzing the intersection scene, the automatic driving vehicle can predict the possible traffic situation, take safety measures in advance, and improve the safety of the automatic driving vehicle.

[0059] Currently, the related technology adopts high-precision maps for intersection construction when constructing intersections. However, there are the following disadvantages in constructing intersections based on high-precision maps: 1. The cost of high-precision map making is high, which consumes a large amount of human, material and financial resources; 2. The accuracy of high-precision maps is limited by the speed of geographical data collection and updating, for example, some newly built roads may not be marked on the high-precision map; 3. In some sudden situations such as road condition changes, the road condition information on the high-precision map cannot be updated in time, which may cause navigation errors; 4. The precision of high-precision maps is limited by the precision of collection equipment and manual processing precision, and the data is not accurate enough; 5. High-precision map updating requires certain time and cost, and the high-precision map of some areas is not updated in time.

[0060] Based on the above problems existing in the related art, the application provides a method for constructing an intersection, which comprises the following steps: acquiring initial intersection information of a to-be-constructed intersection corresponding to a vehicle; acquiring real-time perception data of the to-be-constructed intersection corresponding to the vehicle based on collected data of a plurality of sensors of the vehicle; determining intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data; and optimizing the initial intersection information based on the intersection road line prediction information to obtain target intersection information. The initial intersection information is corrected by the perception data corresponding to the real-time collected data of the plurality of sensors of the vehicle, and the construction of the intersection is realized. Compared with the construction of the intersection by using a high-precision map, the accuracy of the construction of the intersection is improved and the cost of the construction of the intersection is reduced by using the real-time perception data of the vehicle.

[0061] As shown in FIG. 1, FIG. 1 is a structural schematic diagram of an intersection construction device in a hardware running environment related to an embodiment scheme of the application.

[0062] The intersection construction device in the embodiment of the application can be a vehicle. As shown in FIG. 1, the intersection construction device can comprise a processor 1001 such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication among the components. The user interface 1003 can comprise a display screen (Display) and an input unit such as a keyboard (Keyboard). Optionally, the user interface 1003 can further comprise a standard wired interface and a wireless interface. The network interface 1004 can optionally comprise a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory) such as a disk memory. The memory 1005 can optionally be a storage device independent of the aforementioned processor 1001.

[0063] Optionally, the intersection construction device can further comprise a camera, an RF (Radio Frequency, radio frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. The sensor can be, for example, a light sensor, a motion sensor, and other sensors. Of course, the intersection construction device can further be configured with a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, and other sensors, which are not described herein again.

[0064] Those skilled in the art can understand that the terminal structure shown in FIG. 1 does not constitute a limitation on the intersection construction device, and can comprise more or fewer components than those shown, or combine certain components, or different component arrangements.

[0065] As shown in FIG. 1, the memory 1005 as a computer storage medium can include an operating system, a network communication module, a user interface module, and an intersection construction program.

[0066] In the intersection construction device shown in FIG. 1, the network interface 1004 is mainly used for connecting a background server and communicating data with the background server; the user interface 1003 is mainly used for connecting a client (user end) and communicating data with the client; and the processor 1001 can be used for calling the intersection construction program stored in the memory 1005.

[0067] In the embodiment, the intersection construction device includes a memory 1005, a processor 1001, and an intersection construction program stored in the memory 1005 and executable on the processor 1001, wherein the processor 1001 calls the intersection construction program stored in the memory 1005 and executes the steps of the intersection construction method in each embodiment.

[0068] The present application also provides an intersection construction method, as shown in FIG. 2, which is a flowchart of a first embodiment of the intersection construction method of the present application.

[0069] The intersection construction method includes:

[0070] Step S101: obtaining initial intersection information of a to-be-constructed intersection corresponding to a vehicle;

[0071] Step S102: obtaining real-time perception data of the to-be-constructed intersection corresponding to the vehicle based on collected data of multiple sensors of the vehicle;

[0072] Step S103: determining intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data;

[0073] Step S104: optimizing the initial intersection information based on the intersection road line prediction information to obtain target intersection information.

[0074] In the process of vehicle driving, for example, in the process of automatic driving, whether the vehicle is about to enter an intersection can be determined in real time by using perception data or a navigation map to identify whether the vehicle is about to enter an intersection. In an embodiment, whether the vehicle is about to enter an intersection can be determined by using identification information of a road sign in the perception data, for example, if the identification information of the road sign includes direction information of multiple driving directions (for example, straight, left turn, and right turn), or the identification information of the road sign includes road information of other roads in the front intersection, it can be determined that the vehicle is about to enter an intersection; or the current position of the vehicle and the navigation map can be used to accurately identify whether the vehicle is about to enter an intersection.

[0075] When the vehicle is about to enter the intersection, initial intersection information of the intersection to be constructed corresponding to the vehicle is obtained, that is, the intersection to be constructed is the intersection that the vehicle is about to enter. In an embodiment, if it is determined that the vehicle is about to enter the intersection, the intersection to be constructed is initialized, and the initialized intersection to be constructed is updated based on the zebra crossing, traffic light and stop line in the perception data to obtain the initial intersection information. For example, the initialized intersection to be constructed is a rectangular frame, and the upper, lower, left and right of the rectangular frame are provided with a plurality of entrances and exits. The vehicle obtains information such as the zebra crossing, traffic light and stop line corresponding to the intersection to be constructed through the perception data, and updates the initialized intersection to be constructed according to the zebra crossing, traffic light and stop line, that is, determines whether the entrance and exit in the rectangular frame is a real entrance and exit through the zebra crossing, traffic light and stop line, and updates the entrance and exit of the rectangular frame (the entrance and exit in the initialized intersection to be constructed) based on the real entrance and exit to obtain the initial intersection information. For example, for the real entrance and exit in the entrance and exit of the rectangular frame, the zebra crossing or stop line corresponding to the real entrance and exit is updated in the rectangular frame, and for the non-real entrance and exit in the entrance and exit of the rectangular frame, the non-real entrance and exit is deleted in the entrance and exit of the rectangular frame, and finally the initial intersection information is formed.

[0076] It should be noted that in a possible implementation, the service platform can pre-store preset initial intersection information of each intersection in each road. The vehicle can send an intersection obtaining request to the service platform based on the current position and driving direction of the vehicle. The service platform queries corresponding initial intersection information from the plurality of preset initial intersection information according to the current position and driving direction of the vehicle in the intersection obtaining request, and feeds back the initial intersection information to the vehicle. Alternatively, the service platform can pre-store preset intersection construction information of each intersection in each road. The preset intersection construction information can include the design scale of each intersection and traffic light signal information. The vehicle can send an information obtaining request to the service platform based on the current position and driving direction of the vehicle. The service platform queries corresponding intersection construction information from the plurality of preset intersection construction information according to the current position and driving direction of the vehicle in the information obtaining request, and feeds back the intersection construction information to the vehicle. The vehicle constructs initial intersection information according to the received intersection construction information.

[0077] It can be understood that when obtaining the initial intersection information, the vehicle can first construct the initial intersection information based on the perception data. If the effect of the initial intersection information is not good, for example, the initial intersection information constructed by the perception data cannot cover the overall situation of the intersection to be constructed, for example, the difference between the number of entrances and exits of the initial intersection information generated by the perception data and the number of real entrances and exits of the intersection to be constructed is greater than a preset value, the initial intersection information of the intersection to be constructed is obtained through the service platform.

[0078] It should be noted that, according to the historical driving track of the vehicle, the preset initial intersection information corresponding to each intersection can also be pre-stored in the vehicle, for example, the preset initial intersection information can include the initial intersection information of all intersections within the preset distance corresponding to the historical driving track, and the vehicle can query the initial intersection information in the pre-stored preset initial intersection information based on the current position and the driving direction of the vehicle. Of course, the preset intersection construction information of each intersection can also be pre-stored, and the vehicle can query the corresponding intersection construction information according to the current position and the driving direction of the vehicle.

[0079] Further, in a possible implementation, before step S101, the intersection construction method further includes:

[0080] In step S105, the design scale corresponding to the to-be-constructed intersection and the traffic light signal information are acquired.

[0081] In step S106, the initial intersection information corresponding to the to-be-constructed intersection is determined based on the design scale and the traffic light signal information.

[0082] Before the intersection construction, the design scale corresponding to the to-be-constructed intersection and the traffic light signal information can be acquired, wherein the design scale refers to the design size of the to-be-constructed intersection, and the traffic light signal information can be the position of the traffic light in the to-be-constructed intersection. Then, the initial intersection information corresponding to the to-be-constructed intersection is determined based on the design scale and the traffic light signal information. First, four directions are fixed in the intersection model, and each direction has an exit and an entrance to form a set of exits and entrances. The size of the exit and entrance can be assumed to be a fixed value first, and the position of the exit and entrance of each direction is adjusted according to the traffic light signal information, for example, the exit and entrance are increased or the position of the exit and entrance is adjusted. The position of the exit and entrance of each direction can be adjusted according to the stop line / zebra crossing information in the to-be-constructed intersection, wherein there can be multiple sets of exits and entrances in a certain direction. The initial intersection information corresponding to the to-be-constructed intersection is obtained, wherein the initial intersection information can include the initial intersection size, the exit and entrance position, the exit and entrance direction (driving direction), the number of exits and entrances, etc.

[0083] The vehicle is provided with a laser radar and a plurality of cameras and other sensors, and the vehicle collects data in real time through the sensors during driving. An upstream perception module of the intersection construction in the vehicle processes the collected data to obtain perception data, and the vehicle can directly acquire real-time perception data of the to-be-constructed intersection from the upstream perception module. The real-time perception data can include vehicle information, signal light state information, pedestrian information, vehicle lane, sidewalk, zebra crossing, obstacle information, etc.

[0084] After obtaining the real-time perception data, the intersection road line prediction information corresponding to the to-be-constructed intersection is determined based on the real-time perception data. In an embodiment, after obtaining the real-time perception data, the real-time perception data is filtered to filter out data irrelevant to the to-be-constructed intersection, to obtain target perception data corresponding to the to-be-constructed intersection, and the intersection road line prediction information is determined according to the target perception data, wherein the intersection road line prediction information can include a stop line, a zebra crossing, and a pedestrian crossing area. Further, in a possible implementation, step S103 includes:

[0085] In step S1031, the real-time perception data is filtered based on the initial intersection information to obtain target perception data.

[0086] In step S1032, the intersection road line prediction information corresponding to the to-be-constructed intersection is determined based on the target perception data.

[0087] After obtaining the real-time perception data, the real-time perception data is filtered based on the initial intersection information of the to-be-constructed intersection to filter out data irrelevant to the to-be-constructed intersection, to obtain target perception data. Further, in a possible implementation, step S1031 includes:

[0088] In step a, the initial intersection size corresponding to the initial intersection information is obtained.

[0089] In step b, the real-time perception data is filtered based on the initial intersection size to obtain the target perception data.

[0090] After obtaining the real-time perception data, the initial intersection size of the initial intersection information is obtained, the initial intersection size can be the road area size occupied by the initial intersection, and the real-time perception data is filtered based on the initial intersection size to filter out data irrelevant to the to-be-constructed intersection, to obtain target perception data, wherein the target perception data is data in the initial intersection in the real-time perception data, to reduce the influence of data outside the initial intersection on intersection construction, and to improve the accuracy of intersection construction.

[0091] After obtaining the target perception data, the intersection road line prediction information corresponding to the to-be-constructed intersection is determined based on the target perception data. In an embodiment, the road lines in the target perception data are clustered, and the stop line, the zebra crossing, and the pedestrian crossing area in the intersection road line prediction information are determined according to the clustering result, so that the intersection road line prediction information in the intersection can be accurately obtained, and the accuracy of intersection construction is further improved.

[0092] After the intersection road line prediction information is obtained, the initial intersection information is optimized based on the intersection road line prediction information to obtain target intersection information. In an embodiment, the initial intersection information can be optimized based on the intersection road line prediction information to obtain optimized initial intersection information, for example, the entrance and exit positions in the initial intersection information are optimized based on the stop line, zebra crossing or pedestrian crossing area in the intersection road line prediction information, and the target intersection information is obtained based on the optimized entrance and exit positions.

[0093] By obtaining the initial intersection information of the intersection corresponding to the vehicle; then obtaining the real-time perception data of the intersection corresponding to the vehicle based on the collection data of the plurality of sensors of the vehicle; then determining the intersection road line prediction information corresponding to the intersection to be constructed based on the real-time perception data; and then optimizing the initial intersection information based on the intersection road line prediction information to obtain the target intersection information. The application can optimize the initial intersection information in real time through the perception data of the vehicle, which can reduce the influence of obstacles and vehicles on the intersection construction compared with the intersection construction using high-precision maps, improve the accuracy of intersection construction, improve the passing rate and safety of the autonomous vehicle at the intersection, and reduce the cost of intersection construction.

[0094] Based on the first embodiment, a second embodiment of the intersection construction method of the application is proposed, wherein step S1032 comprises:

[0095] Step S201, optimizing the initial intersection information based on the intersection road line prediction information to obtain optimized initial intersection information;

[0096] Step S202, adjusting the optimized initial intersection information based on the vehicle information, signal light state information and obstacle information in the target perception data to obtain the target intersection information.

[0097] After the intersection road line prediction information is obtained, the initial intersection information is optimized based on the intersection road line prediction information to obtain optimized initial intersection information, for example, the entrance and exit positions corresponding to the intersection to be constructed can be determined based on the intersection road line prediction information, and the initial intersection information is optimized based on the entrance and exit positions. In a possible implementation, step S201 comprises:

[0098] Step S2011, determining the entrance and exit positions corresponding to the intersection to be constructed based on the intersection road line prediction information;

[0099] Step S2012, optimizing the initial intersection information based on the entrance and exit positions to obtain optimized initial intersection information.

[0100] After obtaining the intersection road line prediction information, an entrance and exit position corresponding to the intersection to be constructed is determined according to the intersection road line prediction information; further, in a possible implementation manner, step S2011 comprises:

[0101] Step c, determining the entrance and exit position corresponding to the intersection to be constructed based on the stop line in the intersection road line prediction information;

[0102] Step d, if there is no stop line in the intersection road line prediction information, determining the entrance and exit position corresponding to the intersection to be constructed based on the zebra crossing in the intersection road line prediction information;

[0103] Step e, if there is no zebra crossing in the intersection road line prediction information, clustering the target perception information, and determining the entrance and exit position corresponding to the intersection to be constructed based on the clustering result.

[0104] After obtaining the intersection road line prediction information, the stop line in the intersection road line prediction information is obtained, if there is a stop line in the intersection road line prediction information, the entrance and exit position corresponding to the intersection to be constructed is determined according to the stop line, and specifically, the entrance and exit position corresponding to the intersection to be constructed is determined according to the area formed by each stop line. If there is no stop line in the intersection road line prediction information, the zebra crossing in the intersection road line prediction information is obtained, if there is a zebra crossing in the intersection road line prediction information, the entrance and exit position corresponding to the intersection to be constructed is determined according to the area formed by each zebra crossing, that is, the area formed by the zebra crossing is a pedestrian passing area, and the entrance and exit position corresponding to the intersection to be constructed can be determined according to the area surrounded by the pedestrian passing areas in multiple directions. If there is no zebra crossing and stop line in the intersection road line prediction information, the target perception information is clustered, and the entrance and exit position corresponding to the intersection to be constructed is determined based on the clustering result. In an embodiment, the pedestrian information in the target perception information is obtained, the k-means clustering algorithm is used to cluster the pedestrian information, the pedestrian passing area is obtained according to the clustering result, and the entrance and exit position corresponding to the intersection to be constructed can be determined according to the area surrounded by the pedestrian passing areas in multiple directions.

[0105] Through the stop line, the zebra crossing and the clustering result of the target perception information, the entrance and exit position corresponding to the intersection to be constructed can be accurately determined, so as to optimize the initial intersection information according to the entrance and exit position corresponding to the intersection to be constructed, and improve the accuracy of intersection construction.

[0106] After determining the entrance and exit position corresponding to the intersection to be constructed, the initial intersection information is optimized based on the entrance and exit position, that is, the entrance and exit position in the initial intersection information is corrected according to the entrance and exit position corresponding to the intersection to be constructed, and the optimized initial intersection information is obtained.

[0107] After obtaining the optimized initial intersection information, vehicle information, signal light state information and obstacle information in the target perception data are obtained, and the optimized initial intersection information is adjusted based on the vehicle information, signal light state information and obstacle information to obtain the target intersection information. In a possible implementation manner, step S202 includes:

[0108] In step S2021, the access information in the optimized initial intersection information is adjusted based on the signal light state information and the obstacle information.

[0109] In step S2022, the access direction in the optimized initial intersection information is determined based on the route arrow information and the traffic information corresponding to the vehicle information.

[0110] After obtaining the vehicle information, signal light state information and obstacle information in the target perception data, the access information in the optimized initial intersection information is adjusted based on the signal light state information and the obstacle information. In an embodiment, the intersection scene map can be generated according to the optimized initial intersection information, the intersection scene map is updated based on the vehicle information, signal light state information and obstacle information in the target perception data, that is, the vehicle information, signal light state information and obstacle information in the target perception data are added to the intersection scene map, for example, the vehicle, signal light and obstacle are added to the intersection scene map according to the vehicle position corresponding to the vehicle information, the signal light position corresponding to the signal light state information and the obstacle position corresponding to the obstacle information, so as to reflect the vehicle, signal light and obstacle corresponding to the target perception data in the intersection scene map, and the access information in the optimized initial intersection information is adjusted based on the intersection scene map, for example, the position or size of the access is adjusted, the corresponding access is deleted when the access is occupied by the obstacle, or the access size of the corresponding access is adjusted when the access is partially occupied by the obstacle, the access direction of the corresponding access is adjusted according to the vehicle or signal light in the intersection scene map, and the like.

[0111] Then, the route arrow information in the target perception data and the traffic information corresponding to the vehicle information in the target perception data can also be obtained, the access direction in the optimized initial intersection information is determined according to the route arrow information and the traffic information, for example, the access direction of the corresponding access is determined according to the traffic information or the route arrow information, and the adjusted access direction matches the traffic direction of the traffic information or the route arrow information. After adjusting the access information and the access direction in the optimized initial intersection information, the target intersection information is obtained to improve the accuracy of the target intersection information.

[0112] The application optimizes the initial intersection information based on the intersection road line prediction information, obtains optimized initial intersection information, and then adjusts the optimized initial intersection information based on vehicle information, signal light state information and obstacle information in the target perception data, to obtain the target intersection information. The application optimizes the initial intersection, improves the accuracy of the target intersection information, and further improves the accuracy of the intersection construction, the passing rate and safety of the autonomous vehicle at the intersection.

[0113] Based on the first embodiment, a third embodiment of the intersection construction method of the application is provided, wherein after step S104, the intersection construction method further comprises:

[0114] In step S301, the intersection obstacle information and the abnormal vehicle information corresponding to the target intersection information are determined based on the real-time perception data.

[0115] In step S302, the drivable area corresponding to the vehicle is determined based on the target intersection information, the intersection obstacle information and the abnormal vehicle information.

[0116] After obtaining the target intersection information, the intersection obstacle information and the abnormal vehicle information corresponding to the target intersection information are determined based on the real-time perception data. In an embodiment, the vehicle information and obstacle information in the real-time perception data within the area corresponding to the target intersection information are obtained, and the abnormal vehicle information of the abnormal driving vehicle, the sudden navigation vehicle and the emergency vehicle in the traffic flow information corresponding to the target intersection information is determined.

[0117] After obtaining the abnormal vehicle information, the drivable area corresponding to the vehicle is determined based on the target intersection information, the intersection obstacle information and the abnormal vehicle information. Specifically, the drivable area is determined according to the driving direction of the vehicle, the intersection obstacle information and the abnormal vehicle information in the area corresponding to the target intersection information, so that the autonomous vehicle drives based on the drivable area.

[0118] In a possible implementation, after step S302, the intersection construction method further comprises:

[0119] In step S303, the drivable area is optimized based on the pedestrian information and the signal light state information in the real-time perception data, to obtain the target drivable area.

[0120] After the drivable area is obtained, vehicle information, pedestrian information and signal light state information in the real-time perception data can be obtained, pedestrian information and signal light state information in the real-time perception data in the region corresponding to the target intersection information are obtained, the drivable area is optimized based on the region where the pedestrian information and the signal light state information are located, so that the optimized target drivable area does not include the region where the pedestrian information and the signal light state information are located, and the passing rate and safety of the autonomous vehicle at the intersection are further improved.

[0121] The application further improves the passing rate and safety of the autonomous vehicle at the intersection by determining the intersection obstacle information and the abnormal vehicle information corresponding to the target intersection information based on the real-time perception data, and then determining the drivable area corresponding to the vehicle based on the target intersection information, the intersection obstacle information and the abnormal vehicle information.

[0122] In addition, the application also provides a vehicle, as shown in FIG. 3, which comprises:

[0123] The first acquisition module 10 is configured to acquire initial intersection information of a to-be-constructed intersection corresponding to a vehicle.

[0124] The second acquisition module 20 is configured to acquire real-time perception data of the to-be-constructed intersection corresponding to the vehicle based on the collected data of a plurality of sensors of the vehicle.

[0125] The determination module 30 is configured to determine intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data.

[0126] The optimization module 40 is configured to optimize the initial intersection information based on the intersection road line prediction information to obtain target intersection information.

[0127] The determination of the intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data comprises:

[0128] The real-time perception data is filtered based on the initial intersection information to obtain target perception data, and the intersection road line prediction information corresponding to the to-be-constructed intersection is determined based on the target perception data.

[0129] The optimization of the initial intersection information based on the intersection road line prediction information to obtain the target intersection information comprises:

[0130] The initial intersection information is optimized based on the intersection road line prediction information to obtain optimized initial intersection information, and the optimized initial intersection information is adjusted based on vehicle information, signal light state information and obstacle information in the target perception data to obtain the target intersection information.

[0131] The initial intersection information is optimized based on the intersection road line prediction information to obtain optimized initial intersection information, including:

[0132] The exit and entrance positions corresponding to the to-be-constructed intersection are determined based on the intersection road line prediction information, and the exit and entrance positions in the initial intersection information are corrected based on the exit and entrance positions corresponding to the to-be-constructed intersection to obtain the optimized initial intersection information.

[0133] The exit and entrance positions corresponding to the to-be-constructed intersection are determined based on the intersection road line prediction information, including:

[0134] The exit and entrance positions corresponding to the to-be-constructed intersection are determined based on the stop line in the intersection road line prediction information, if there is no stop line in the intersection road line prediction information, the exit and entrance positions corresponding to the to-be-constructed intersection are determined based on the zebra crossing in the intersection road line prediction information, and if there is no zebra crossing in the intersection road line prediction information, the target perception information is clustered, and the exit and entrance positions corresponding to the to-be-constructed intersection are determined based on the clustering result.

[0135] The initial intersection information is optimized based on the intersection road line prediction information to obtain the target intersection information, including:

[0136] The exit and entrance information in the optimized initial intersection information is adjusted based on the signal light state information and the obstacle information, and the exit and entrance direction in the optimized initial intersection information is determined based on the route arrow information and the vehicle flow information corresponding to the vehicle information.

[0137] The real-time perception data is filtered based on the initial intersection information to obtain the target perception data, including:

[0138] The initial intersection size corresponding to the initial intersection information is obtained, and the real-time perception data is filtered based on the initial intersection size to obtain the target perception data.

[0139] After the initial intersection information is optimized based on the intersection road line prediction information to obtain the target intersection information, the following steps are further included:

[0140] The intersection obstacle information and the abnormal vehicle information corresponding to the target intersection information are determined based on the real-time perception data, and the drivable area corresponding to the vehicle is determined based on the target intersection information, the intersection obstacle information and the abnormal vehicle information.

[0141] After the drivable area corresponding to the vehicle is determined based on the target intersection information, the intersection obstacle information and the abnormal vehicle information, the following steps are further included:

[0142] optimizing the drivable area based on the pedestrian information and the signal light state information in the real-time perception data, to obtain a target drivable area.

[0143] Before the obtaining of the initial intersection information of the intersection to be constructed corresponding to the vehicle, the method further comprises:

[0144] obtaining design dimensions and traffic light signal information corresponding to the intersection to be constructed, and determining the initial intersection information corresponding to the intersection to be constructed based on the design dimensions and the traffic light signal information.

[0145] The obtaining of the initial intersection information of the intersection to be constructed corresponding to the vehicle comprises:

[0146] sending an intersection obtaining request to a service platform based on the current position and the driving direction of the vehicle, and receiving initial intersection information fed back by the service platform, wherein the service platform obtains the initial intersection information corresponding to the intersection obtaining request based on the current position and the driving direction of the vehicle in a plurality of preset initial intersection information queries.

[0147] The method performed by each program unit can refer to each embodiment of the intersection construction method of the present application, and will not be described here.

[0148] In addition, an embodiment of the present application further proposes a computer readable storage medium, and the computer readable storage medium stores an intersection construction program, and the intersection construction program is executed by a processor to implement the steps of the intersection construction method.

[0149] In addition, an embodiment of the present application further proposes a computer program product, and the computer program product comprises an intersection construction program, and the intersection construction program is executed by a processor to implement the steps of the intersection construction method.

[0150] It should be noted that in this document, the term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or system. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or system including the element.

[0151] The above sequence numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0152] Those skilled in the art can clearly understand the above-mentioned embodiment method can be realized by means of software and necessary general hardware platform, of course, also can be through hardware, but in many cases the former is the better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of software product, and the computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, including a plurality of instructions to make a terminal device (may be a mobile phone, computer, server, air conditioner, or network equipment, etc.) execute the method described in various embodiments of the present application.

[0153] The above has described the embodiments of the present application, the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical application or improvement of the technology in the market, or to enable other ordinary skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method of constructing an intersection, characterized by The method comprises the following steps: acquiring initial intersection information of a to-be-constructed intersection corresponding to a vehicle; acquiring real-time perception data of the to-be-constructed intersection corresponding to the vehicle based on collected data of a plurality of sensors of the vehicle; determining intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data; optimizing the initial intersection information based on the intersection road line prediction information to obtain target intersection information.

2. The intersection construction method of claim 1, wherein, The step of determining intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data comprises the following steps: performing filtering operation on the real-time perception data based on the initial intersection information to obtain target perception data; determining intersection road line prediction information corresponding to the to-be-constructed intersection based on the target perception data.

3. The intersection construction method of claim 2, wherein, The step of optimizing the initial intersection information based on the intersection road line prediction information to obtain target intersection information comprises the following steps: optimizing the initial intersection information based on the intersection road line prediction information to obtain optimized initial intersection information; adjusting the optimized initial intersection information based on vehicle information, signal light state information and obstacle information in the target perception data to obtain the target intersection information.

4. The intersection construction method of claim 3, wherein, The step of optimizing the initial intersection information based on the intersection road line prediction information to obtain optimized initial intersection information comprises the following steps: determining entrance and exit positions corresponding to the to-be-constructed intersection based on the intersection road line prediction information; correcting the entrance and exit positions in the initial intersection information based on the entrance and exit positions corresponding to the to-be-constructed intersection to obtain the optimized initial intersection information.

5. The intersection construction method of claim 4, wherein, The step of determining entrance and exit positions corresponding to the to-be-constructed intersection based on the intersection road line prediction information comprises the following steps: determining entrance and exit positions corresponding to the to-be-constructed intersection based on stop lines in the intersection road line prediction information; if there is no stop line in the intersection road line prediction information, determining entrance and exit positions corresponding to the to-be-constructed intersection based on zebra crossings in the intersection road line prediction information; if there is no zebra crossing in the intersection road line prediction information, performing clustering on the target perception information, and determining entrance and exit positions corresponding to the to-be-constructed intersection based on clustering results.

6. The intersection construction method of claim 3, wherein, The step of adjusting the optimized initial intersection information based on vehicle information, signal light state information and obstacle information in the target perception data to obtain the target intersection information comprises the following steps: adjusting entrance and exit information in the optimized initial intersection information based on the signal light state information and the obstacle information; determining entrance and exit directions in the optimized initial intersection information based on route arrow information and vehicle flow information corresponding to vehicle information.

7. The intersection construction method of claim 2, wherein, The step of performing filtering operation on the real-time perception data based on the initial intersection information to obtain target perception data comprises the following steps: acquiring an initial intersection size corresponding to the initial intersection information; performing filtering operation on the real-time perception data based on the initial intersection size to obtain the target perception data.

8. The intersection construction method of claim 1, wherein, The method further comprises the following steps after the step of optimizing the initial intersection information based on the intersection road line prediction information to obtain target intersection information: determine intersection obstacle information corresponding to the target intersection information and abnormal vehicle information based on the real-time perception data; determine a drivable area corresponding to the vehicle based on the target intersection information, the intersection obstacle information, and the abnormal vehicle information.

9. The intersection construction method of claim 8, wherein, The step of determining a drivable area corresponding to the vehicle based on the target intersection information, the intersection obstacle information, and the abnormal vehicle information further comprises: optimizing the drivable area based on pedestrian information and signal light state information in the real-time perception data to obtain a target drivable area.

10. The intersection construction method according to any one of claims 1 to 9, characterized by, The step of obtaining initial intersection information of a to-be-constructed intersection corresponding to the vehicle further comprises: obtain design dimensions and traffic light signal information corresponding to the to-be-constructed intersection; determine initial intersection information corresponding to the to-be-constructed intersection based on the design dimensions and the traffic light signal information.

11. The intersection construction method according to any one of claims 1 to 9, characterized by, The step of obtaining initial intersection information of a to-be-constructed intersection corresponding to the vehicle comprises: send an intersection obtaining request to a service platform based on the current position and the driving direction of the vehicle, and receive initial intersection information fed back by the service platform, wherein the service platform obtains corresponding initial intersection information from a plurality of preset initial intersection information queries based on the current position and the driving direction of the vehicle in the intersection obtaining request.

12. A vehicle characterized by comprising: The vehicle comprises: a first obtaining module configured to obtain initial intersection information of a to-be-constructed intersection corresponding to the vehicle; a second obtaining module configured to obtain real-time perception data of the to-be-constructed intersection corresponding to the vehicle based on collection data of a plurality of sensors of the vehicle; a determining module configured to determine intersection road line prediction information corresponding to the to-be-constructed intersection based on the real-time perception data; an optimization module configured to optimize the initial intersection information based on the intersection road line prediction information to obtain target intersection information.

13. An intersection construction device, characterized by The intersection construction device comprises a memory, a processor, and an intersection construction program stored on the memory and executable on the processor, and the intersection construction program, when executed by the processor, implements the steps of the intersection construction method according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that, The computer readable storage medium stores an intersection construction program, and the intersection construction program, when executed by the processor, implements the steps of the intersection construction method according to any one of claims 1 to 11.

15. A computer program product, characterised in that, The computer program product comprises an intersection construction program, and the intersection construction program, when executed by the processor, implements the steps of the intersection construction method according to any one of claims 1 to 11.

Images