Vehicle driving control method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a vehicle driving control method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining environment perception information, vehicle state information and decision planning information of a target vehicle; constructing a target lane model based on the environment perception information, the vehicle state information and the decision planning information; determining obstacle semantic information and road right relation information based on the target lane model; and controlling the target vehicle to make a driving decision according to the target lane model, the obstacle semantic information and the road right relation information. The vehicle driving control method solves the technical problem that a vehicle driving control method provided in the related technology is low in accuracy and stability.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of intelligent driving technology, and in particular to a vehicle driving control method, device, electronic device and storage medium. Background Art

[0002] At present, the establishment of perception environment model, lane model and semantic information mainly adopts the following two methods: lane line fitting method based on polynomial regression and lane line detection method based on key points. Among them, the lane line fitting method based on polynomial regression constructs the lane model by fitting the points on the lane line to the polynomial curve; the lane line detection method based on key points focuses on identifying the key feature points on the lane line to construct the lane line model. However, both of the above methods do not fully consider the impact of noise factors such as positioning jitter and obstacles on the establishment of lane models. At the same time, when dealing with the conflict relationship between obstacles and target vehicles, the relevant technology lacks an effective integration mechanism, and fails to integrate the predicted behavior of obstacles and their road rights relationship relative to the target vehicle into the lane semantic information, which will reduce the accuracy of the final output lane model and semantic information results, affect the accuracy of downstream module decision-making, and thus affect the safety of intelligent driving vehicles and systems. Summary of the invention

[0003] The embodiments of the present application provide a vehicle driving control method, device, electronic device and storage medium, aiming to improve the technical problems of low accuracy and stability in the vehicle driving control method provided in the related art.

[0004] According to one embodiment of the present application, a vehicle driving control method is provided, including: obtaining environmental perception information, vehicle status information and decision-making planning information of a target vehicle; constructing a target lane model based on the environmental perception information, vehicle status information and decision-making planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment where the target vehicle is located; determining obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle; and controlling the target vehicle to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information.

[0005] The above optional embodiments of the present application can achieve the following beneficial effects: first, the environmental perception information, vehicle status information and decision-making planning information of the target vehicle are obtained, and a target lane model is constructed based on the above information; then, obstacle semantic information and road right relationship information are determined based on the target lane model; finally, the target vehicle is controlled to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information, thereby achieving the purpose of improving the environmental model of the vehicle driving control system and enhancing the accuracy and comprehensiveness of the basis for driving decisions, thereby achieving the technical effect of improving the accuracy and stability of vehicle driving control, and then solving the technical problem of low accuracy and stability of the vehicle driving control method provided in the relevant technology.

[0006] Optionally, constructing a target lane model based on environmental perception information, vehicle status information and decision planning information includes: performing time synchronization processing on the environmental perception information, vehicle status information and decision planning information to obtain a time synchronization result; performing combination processing on the time synchronization result to obtain a lane list combination; performing conversion processing on the lane list combination to obtain an initial lane model; and performing correction processing on the initial lane model to obtain a target lane model.

[0007] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: through time synchronization processing, the time offset of data between different sensors and modules can be eliminated, and the synchronization of environmental perception information, vehicle status information and decision-making planning information can be ensured; the time-synchronized information is combined to form a lane list combination, which provides a complete lane information framework for subsequent modeling; further, through the conversion processing of the lane list combination, the abstract lane information is converted into a visual lane model, and a digital representation of the vehicle driving environment is preliminarily established; finally, through correction processing, a target lane model that accurately reflects the vehicle driving environment can be constructed to ensure that the autonomous driving vehicle can make decisions based on accurate and real-time environmental information when performing actions such as lane changing and obstacle avoidance, thereby improving the safety of vehicle driving and the smoothness of the driving experience.

[0008] Optionally, the initial lane model is corrected to obtain a target lane model, including: acquiring historical planned lane information and current detected lane information based on the initial lane model, wherein the historical planned lane information includes: a reference lane list and a target lane list corresponding to the historical data frame, the reference lanes in the reference lane list are used to represent the decision driving lanes at the historical decision moments, the target lanes in the target lane list are used to represent the expected driving lanes at the historical decision moments, and the current detected lane information includes: the current lane corresponding to the current data frame; matching processing is performed using the historical planned lane information and the current detected lane information to obtain a lane matching result, wherein the lane matching result is used to determine whether the target vehicle has performed lateral behavior; in response to determining that the historical planned lane information is inconsistent with the current detected lane information based on the lane matching result, the current detected lane information in the initial lane model is corrected to obtain the target lane model.

[0009] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: by real-time analysis of the differences between historical planned lane information and current detected lane information, it is possible to timely identify and correct lateral behavior deviations that may occur in the target vehicle during driving, ensuring that the lane model matches the actual driving state of the vehicle, thereby improving the accuracy and real-time performance of the vehicle driving control system environmental model.

[0010] Optionally, the current detected lane information in the initial lane model is corrected to obtain the target lane model, including: obtaining the lateral distance between the target vehicle and the target lane; in response to the lateral distance being less than a preset distance, correcting the current lane in the initial lane model to the target lane to obtain the target lane model, wherein the preset distance is determined based on the lane width of the target lane.

[0011] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: by real-time monitoring and calculating the lateral distance between the target vehicle and the center line of the target lane, and comparing it with the preset distance, it can timely identify whether the target vehicle has successfully transitioned to the target lane, ensuring that the vehicle driving control system can dynamically adjust the lane model according to the actual driving state of the vehicle, avoiding unnecessary lane information correction or lane changing operations caused by positioning errors or system noise, thereby improving the accuracy of the lane model and the real-time response of the system.

[0012] Optionally, determining obstacle semantic information and road right relationship information based on the target lane model includes: acquiring obstacle path information and obstacle prediction information corresponding to the target obstacle based on environmental perception information; projecting the target obstacle to the target lane model based on the obstacle path information and obstacle prediction information to obtain obstacle semantic information; determining a driving conflict type based on the obstacle semantic information, wherein the driving conflict type includes at least one of the following: lane keeping type, merging type, cutting-in type, and intersection conflict type; and determining road right relationship information based on the driving conflict type.

[0013] The above optional embodiments of the present application can achieve the following beneficial effects: by comprehensively analyzing environmental perception information, obstacle path information, obstacle prediction information and target lane model, real-time and detailed obstacle semantic information can be constructed, and then the type of driving conflict between the target vehicle and the obstacle can be accurately determined, ensuring that the vehicle driving control system can not only instantly identify obstacles in the surrounding environment, but also determine the road right relationship information according to the different driving conflict types of obstacles, thereby providing key basis for subsequent decision-making.

[0014] Optionally, determining the driving conflict type based on obstacle semantic information includes: acquiring an obstacle lane sequence and a reference line lane sequence; determining an obstacle current lane corresponding to a target obstacle based on the obstacle semantic information; and in response to overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane belongs to the reference line lane sequence, determining that the driving conflict type is a lane keeping type.

[0015] The above optional embodiment of the present application can achieve the following beneficial effects: when there is an overlap between the obstacle lane sequence and the reference line lane sequence, and the current lane of the obstacle belongs to the reference line lane sequence, the driving conflict type can be determined to be a lane keeping type. By comparing the obstacle lane sequence of the target vehicle with the reference line lane sequence in real time, the conflict type between the vehicle and the obstacle can be accurately identified and determined, thereby improving overall driving safety and efficiency.

[0016] Optionally, the vehicle driving control method further includes: in response to the absence of overlap between the obstacle lane sequence and the reference line lane sequence, and the target vehicle and the target obstacle satisfying preset position conditions, determining that the driving conflict type is an intersection conflict type.

[0017] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: by real-time analysis and prediction of the possibility of future intersection between the target vehicle and the obstacle, it can accurately determine whether the driving conflict type between the two is an intersection conflict type under the condition that the obstacle lane sequence and the reference line lane sequence do not overlap, so that the vehicle driving control system can identify and deal with potential intersection conflicts in advance in complex road environments, such as intersections, multi-lane intersections and other scenarios, so that more timely and effective avoidance or adjustment measures can be taken to avoid collision risks and ensure driving safety and road traffic efficiency.

[0018] Optionally, determining the driving conflict type based on obstacle semantic information includes: acquiring an obstacle lane sequence, a reference line lane sequence and a driving behavior prediction result; determining an obstacle current lane corresponding to a target obstacle based on the obstacle semantic information; in response to overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane does not belong to the reference line lane sequence, determining the driving conflict type based on the overlapping lane type and the driving behavior prediction result.

[0019] The above optional embodiments of the present application can achieve the following beneficial effects: by real-time analysis of the relative position of the target vehicle and the obstacle, the driving behavior prediction results of the target vehicle, and the type of overlapping lanes between the reference line lane sequence and the obstacle lane sequence, the type of driving conflict can be more accurately identified and determined, thereby providing key decision-making basis for the vehicle driving control system and ensuring the safe driving of the vehicle under complex road conditions.

[0020] Optionally, the driving conflict type is determined based on the overlapping lane type and the driving behavior prediction results, including: if it is determined based on the driving behavior prediction results that the target vehicle performs lane changing behavior, the driving conflict type is determined to be a cutting-in type; if it is determined based on the driving behavior prediction results that the target vehicle performs tracking behavior, and the overlapping lane type is a forward lane, the driving conflict type is determined to be a merging type; if it is determined based on the driving behavior prediction results that the target vehicle performs tracking behavior, and the overlapping lane type is a backward lane, the driving conflict type is determined to be a lane keeping type.

[0021] The above optional embodiments of the present application can achieve the following beneficial effects: by real-time analysis of the predicted results of the target vehicle's driving behavior and the relative position of the obstacle, combined with the overlap type of the lane sequence planned by the vehicle and the lane where the obstacle is located, the vehicle driving control system can accurately identify and classify the types of driving conflicts, including "cut-in type" conflicts caused by lane changing behavior, "merge type" conflicts caused by front obstacles in tracking behavior, and "lane keeping type" conflicts when rear obstacles approach. It can not only ensure the safe driving of the target vehicle under complex road conditions, but also provide key decision-making basis for the vehicle driving control system to adjust the vehicle's driving strategy, such as timely lane changing, deceleration or acceleration, to avoid collision risks and ensure driving safety.

[0022] Optionally, determining the right-of-way relationship information based on the driving conflict type includes: acquiring the obstacle type of the target obstacle; and determining the right-of-way relationship information based on the driving conflict type, the obstacle type and a preset right-of-way rule.

[0023] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: by real-time monitoring of the type of target obstacle and the type of driving conflict between it and the target vehicle, combined with preset right-of-way rules, the vehicle driving control system can dynamically determine the right-of-way relationship information to ensure that the intelligent driving vehicle can pass safely in various traffic scenarios.

[0024] According to one embodiment of the present application, a vehicle driving control device is also provided, including: an acquisition module, used to acquire environmental perception information, vehicle status information and decision-making planning information of a target vehicle; a construction module, used to construct a target lane model based on the environmental perception information, vehicle status information and decision-making planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment where the target vehicle is located; a determination module, used to determine obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle; a control module, used to control the target vehicle to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information.

[0025] Optionally, the construction module is also used to: perform time synchronization processing on environmental perception information, vehicle status information and decision planning information to obtain a time synchronization result; perform combination processing on the time synchronization result to obtain a lane list combination; perform conversion processing on the lane list combination to obtain an initial lane model; perform correction processing on the initial lane model to obtain a target lane model.

[0026] Optionally, the construction module is also used to: obtain historical planned lane information and current detected lane information based on the initial lane model, wherein the historical planned lane information includes: a reference lane list and a target lane list corresponding to the historical data frame, the reference lane in the reference lane list is used to represent the decision driving lane at the historical decision moment, the target lane in the target lane list is used to represent the expected driving lane at the historical decision moment, and the current detected lane information includes: the current lane corresponding to the current data frame; use the historical planned lane information and the current detected lane information to perform matching processing to obtain a lane matching result, wherein the lane matching result is used to determine whether the target vehicle has lateral behavior; in response to determining that the historical planned lane information is inconsistent with the current detected lane information based on the lane matching result, the current detected lane information in the initial lane model is corrected to obtain a target lane model.

[0027] Optionally, the construction module is also used to: obtain the lateral distance between the target vehicle and the target lane; in response to the lateral distance being less than a preset distance, correct the current lane in the initial lane model to the target lane to obtain a target lane model, wherein the preset distance is determined according to the lane width of the target lane.

[0028] Optionally, the determination module is also used to: obtain obstacle path information and obstacle prediction information corresponding to the target obstacle based on the environmental perception information; project the target obstacle to the target lane model based on the obstacle path information and the obstacle prediction information to obtain obstacle semantic information; determine the driving conflict type based on the obstacle semantic information, wherein the driving conflict type includes at least one of the following: lane keeping type, merging type, cutting-in type, intersection conflict type; determine the road right relationship information based on the driving conflict type.

[0029] Optionally, the determination module is also used to: obtain an obstacle lane sequence and a reference line lane sequence; determine the obstacle current lane corresponding to the target obstacle based on the obstacle semantic information; in response to the overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane belongs to the reference line lane sequence, determine that the driving conflict type is a lane keeping type.

[0030] Optionally, the determination module is further used to: in response to the absence of overlap between the obstacle lane sequence and the reference line lane sequence, and the target vehicle and the target obstacle satisfy preset position conditions, determine that the driving conflict type is an intersection conflict type.

[0031] Optionally, the determination module is also used to: obtain an obstacle lane sequence, a reference line lane sequence and a driving behavior prediction result; determine the obstacle current lane corresponding to the target obstacle based on the obstacle semantic information; in response to overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane does not belong to the reference line lane sequence, determine the driving conflict type based on the overlapping lane type and the driving behavior prediction result.

[0032] Optionally, the determination module is also used for: if it is determined based on the driving behavior prediction results that the target vehicle performs lane changing behavior, determining the driving conflict type as a cut-in type; if it is determined based on the driving behavior prediction results that the target vehicle performs tracking behavior, and the overlapping lane type is a forward lane, determining the driving conflict type as a merging type; if it is determined based on the driving behavior prediction results that the target vehicle performs tracking behavior, and the overlapping lane type is a backward lane, determining the driving conflict type as a lane keeping type.

[0033] Optionally, the determination module is further used to: obtain the obstacle type of the target obstacle; and determine the right-of-way relationship information based on the driving conflict type, the obstacle type and the preset right-of-way rules.

[0034] According to another aspect of an embodiment of the present application, an electronic device is provided, including a processor and a memory, wherein the memory is used to store computer programs; and the processor is used to execute the programs stored in the memory to implement the above-mentioned vehicle driving control method.

[0035] According to another aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the above-mentioned vehicle driving control method is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flow chart of a vehicle driving control method provided by an embodiment of the present application;

[0037] Figure 2 is a schematic diagram of a vehicle driving control system provided by an embodiment of the present application;

[0038] Figure 3 is a schematic diagram of another vehicle driving control method provided by an embodiment of the present application;

[0039] Figure 4 is a schematic diagram of another vehicle driving control method provided by an embodiment of the present application;

[0040] Figure 5 is a schematic diagram of another vehicle driving control method provided by an embodiment of the present application;

[0041] Figure 6is a structural diagram of a vehicle driving control device provided by an embodiment of the present application;

[0042] Figure 7 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] In order to make the technical problems, technical solutions and beneficial effects solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0044] In the related technologies, the lane line fitting method based on polynomial regression and the lane line detection method based on key points have the following limitations in practical applications:

[0045] (1) The impact of positioning jitter is not fully considered: During the driving process of the vehicle, due to the errors of GPS signals, inertial measurement unit (IMU) and other inherent errors of positioning technologies, slight jitter of the vehicle position is inevitable. Positioning jitter will indirectly affect the detection and fitting of lane lines, causing fluctuations in the lane model and affecting its stability. However, when processing lane lines, the above two methods usually only focus on the detection of point cloud data or key points, and do not design an effective mechanism to compensate or correct the model deviation caused by positioning jitter.

[0046] (2) Lack of integration of obstacle prediction behavior and road right relationship: In related technologies, obstacle prediction behavior and road right relationship information are often processed independently, and there is a lack of effective mechanism to integrate them into lane semantic information. As a result, when the downstream decision module processes obstacles, the data source is incomplete or inaccurate, which in turn reduces the reliability and safety of the decision.

[0047] A vehicle driving control method provided in an embodiment of the present application includes: obtaining environmental perception information, vehicle status information and decision-making planning information of a target vehicle; constructing a target lane model based on the environmental perception information, vehicle status information and decision-making planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment where the target vehicle is located; determining obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle; and controlling the target vehicle to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information.

[0048] The above-mentioned vehicle driving control method provided in the embodiment of the present application achieves the following technical effects: first, the environmental perception information, vehicle status information and decision-making planning information of the target vehicle are obtained, and a target lane model is constructed based on the above information; then, the obstacle semantic information and road right relationship information are determined based on the target lane model; finally, the target vehicle is controlled to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information, thereby achieving the purpose of improving the environmental model of the vehicle driving control system and enhancing the accuracy and comprehensiveness of the basis for driving decisions, thereby achieving the technical effect of improving the accuracy and stability of vehicle driving control, and then solving the technical problem of low accuracy and stability of the vehicle driving control method provided in the related art.

[0049] Embodiment 1

[0050] This application embodiment provides a vehicle driving control method, please refer to Figure 1 , including the following steps:

[0051] S11: Acquire environmental perception information, vehicle status information and decision planning information of the target vehicle;

[0052] S12: constructing a target lane model based on the environmental perception information, the vehicle state information and the decision planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment where the target vehicle is located;

[0053] S13: Determine obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle;

[0054] S14: Control the target vehicle to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information.

[0055] The above-mentioned environmental perception information refers to the data about the vehicle's surrounding environment collected by the vehicle driving control system through various sensors, such as lidar, camera, millimeter wave radar, etc. Specifically, environmental perception information includes but is not limited to perception information, positioning information, prediction information, map navigation information, etc.

[0056] The above-mentioned vehicle status information refers to data that describes the status of the target vehicle itself, including but not limited to the vehicle's real-time position, speed, acceleration, direction, attitude angle, wheel speed, steering angle, braking status, accelerator / electric pedal position, engine / motor speed, vehicle health status (such as tire pressure, fuel level, power level) and other information, as well as the status of Advanced Driver Assistance Systems (ADAS), such as the activation status and working status of anti-lock braking system (ABS), traction control system (TCS), blind spot detection (BSD) and other systems.

[0057] The above-mentioned decision planning information refers to the information about vehicle driving strategy and path planning generated by the vehicle decision planning module after the vehicle driving control system receives environmental perception information and vehicle status information. Specifically, the decision planning information includes but is not limited to behavioral decision information, motion planning information, driving behavior prediction results, vehicle control instructions, path feasibility assessment, risk assessment, and dynamic environment adaptation strategy.

[0058] The above obstacle semantic information refers to the information set used in the vehicle driving control system to describe the characteristics and status of obstacles, as well as the relationship between obstacles and vehicle driving. Specifically, obstacle semantic information includes but is not limited to: obstacle type, dynamic attributes, location information, predicted path, etc.

[0059] The above-mentioned road right relationship information refers to the relationship and priority of the target vehicle and other obstacles on the road in terms of right of way, which is determined based on comprehensive factors such as traffic regulations, road rules, and the relative position, dynamic state, predicted path and turning intention of the vehicle and obstacles.

[0060] For example, during the driving process of the target vehicle, the environment perception information around the target vehicle is continuously collected through perception devices such as laser radar, camera and millimeter wave radar. At the same time, based on devices such as acceleration sensor, gyroscope, speed sensor and steering angle sensor, the vehicle driving control system monitors and feeds back its own state information in real time. Subsequently, the vehicle driving control system generates corresponding decision-making planning information based on the acquired environment perception information and vehicle state information, and constructs the target lane model based on the above information, and then determines the obstacle semantic information and road right relationship information based on the constructed target lane model, and finally controls the target vehicle to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information.

[0061] Based on the above steps S11 to S14, first, the environmental perception information, vehicle status information and decision-making planning information of the target vehicle are obtained, and a target lane model is constructed based on the above information; then, the obstacle semantic information and road right relationship information are determined based on the target lane model; finally, the target vehicle is controlled to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information, thereby achieving the purpose of improving the environmental model of the vehicle driving control system and enhancing the accuracy and comprehensiveness of the basis for driving decisions, thereby achieving the technical effect of improving the accuracy and stability of vehicle driving control, and thus solving the technical problem of low accuracy and stability of the vehicle driving control method provided in the related art.

[0062] Figure 2 is a schematic diagram of a vehicle driving control system provided by an embodiment of the present application, such as Figure 2 As shown, the vehicle driving control system in the present application includes the following three parts: an input unit, a cognitive unit and an output unit, wherein the input unit includes the following information: environmental perception information, vehicle status information and decision planning information; the cognitive unit is used for lane information processing, obstacle information processing and lane semantic information processing; the output unit is used to output lane models and environmental information.

[0063] Optionally, in step S12, constructing a target lane model based on the environment perception information, the vehicle state information and the decision planning information includes:

[0064] Step S121, performing time synchronization processing on the environment perception information, the vehicle state information and the decision planning information to obtain a time synchronization result;

[0065] Step S122, combining the time synchronization results to obtain a lane list combination;

[0066] Step S123, converting the lane list combination to obtain an initial lane model;

[0067] Step S124, correcting the initial lane model to obtain a target lane model.

[0068] For example, after receiving the environmental perception information, vehicle status information and decision-making planning information from the input unit, the cognitive unit in the vehicle driving control system will store the above input information in its own data pool. Furthermore, the cognitive unit will perform time synchronization processing on the above input information. Specifically, based on the timestamp of the perception data, the most recent positioning information and map navigation information are matched in the cache queue, thereby achieving time synchronization processing and ensuring the consistency of all input data in the time dimension.

[0069] For example, assume that the timestamp of the current positioning information of the target vehicle is Tloc , the timestamp of the map navigation information is T map , the timestamp of the sensing data is T per , the cognitive unit will look for per The closest time T loc and T map , and determine the difference between the above timestamps, that is, ΔT loc =|T loc -T per | and ΔT map =|T map -T per |, and the relationship between the preset synchronization threshold. If ΔT loc and ΔT map If both are less than the preset synchronization threshold, time synchronization is performed and the timestamps of the positioning information and navigation information are adjusted to T per , forming the time synchronization result. Otherwise, the time synchronization fails, and the latest positioning information and map navigation information are selected and assigned to the decision module for use.

[0070] For prediction information, the cognitive unit will receive the prediction timestamp T pred With T per The difference between ΔT pred As a relative time, it is added to the timestamp of each obstacle prediction point to keep the prediction information relevant to the current moment. If the timestamp of the prediction point is negative, that is, the prediction information is outdated, the cognitive unit will directly discard the prediction point to ensure that only real-time and valid data is processed.

[0071] Furthermore, since the map input information is based on lanes, the attributes and accessories of lanes are described based on lanes. Therefore, after completing the time synchronization process, the time synchronization results need to be combined to obtain a lane list combination. Specifically, the cognitive unit will integrate the attributes and accessories of all lanes to form a lane list combination.

[0072] Exemplarily, the lane list combination may include: Lane A: no obstacles, the vehicle is traveling at a speed of v1, lane width w1, lane line position p1, and a straight lane; Lane B: there is a slow-moving vehicle, speed v2, distance d2, lane width w2, lane line position p2, and a speed limit area; Lane C: there is a vehicle in the right lane ready to cut in, speed v3, direction d3, lane width w3, lane line position p3, and a dynamic obstacle.

[0073] After obtaining the lane list combination, the cognitive unit will convert the abstract lane information into a visual lane model based on the information in the lane list combination, thereby constructing an initial lane model that includes lane attributes and their accessories. Furthermore, the cognitive unit will correct the successfully constructed initial lane model to eliminate errors that may be caused by positioning jitter and obstacle avoidance behavior, thereby obtaining a stable and accurate target lane model.

[0074] Figure 3 is a schematic diagram of another vehicle driving control method provided by an embodiment of the present application, such as Figure 3 As shown, the figure contains three lanes: lane A, lane B and lane C, where lane B is in the intersection area.

[0075] Exemplarily, the starting position of the projection of the intersection area on lane B is close to 0, and the ending position is close to the lane length. When converted to the lane model, the starting position of the intersection area should be calculated based on the current position of the vehicle.

[0076] For example, assuming that the target vehicle is driving on a highway, its vehicle driving control system begins to process the vehicle's environmental perception information, vehicle status information, and decision-making planning information to build an accurate target lane model. First, through time synchronization processing, the environmental perception information obtained by the laser radar, camera, and millimeter-wave radar is aligned with the vehicle's speed, acceleration, and other status information, as well as the driving path and target lane instructions output by the decision-making planning module, to ensure the consistency of all data in the time dimension, thereby obtaining a time synchronization result. Subsequently, the time synchronization result is combined and processed, and the perception data related to the lane information is fused with the decision-making planning data to generate a lane list combination containing all relevant lane information, and the lane list combination is converted and processed, and the data in the lane list combination is used to build an initial lane model. Finally, in order to improve the accuracy of the lane model, the initial lane model is corrected according to the target lane selected in the decision-making planning information and the real-time positioning information of the vehicle to eliminate the errors that may be caused by positioning jitter and obstacle avoidance behavior, thereby obtaining a stable and accurate target lane model.

[0077] Based on the above steps S121 to S124, through time synchronization processing, the time offset of data between different sensors and modules can be eliminated to ensure the synchronization of environmental perception information, vehicle status information and decision-making planning information; the time-synchronized information is combined to form a lane list combination, which provides a complete lane information framework for subsequent modeling; further, through the conversion processing of the lane list combination, the abstract lane information is converted into a visual lane model, and a digital representation of the vehicle driving environment is preliminarily established; finally, through correction processing, a target lane model that accurately reflects the vehicle driving environment can be constructed to ensure that the autonomous driving vehicle can make decisions based on accurate and real-time environmental information when performing actions such as lane changing and obstacle avoidance, thereby improving the safety of vehicle driving and the smoothness of the driving experience.

[0078] Optionally, in step S124, the initial lane model is corrected to obtain a target lane model including:

[0079] Step S1241, acquiring historical planned lane information and current detected lane information based on the initial lane model, wherein the historical planned lane information includes: a reference lane list and a target lane list corresponding to the historical data frame, the reference lane in the reference lane list is used to represent the decision lane at the historical decision moment, the target lane in the target lane list is used to represent the expected lane at the historical decision moment, and the current detected lane information includes: the current lane corresponding to the current data frame;

[0080] Step S1242, performing matching processing using the historical planned lane information and the current detected lane information to obtain a lane matching result, wherein the lane matching result is used to determine whether the target vehicle has a lateral behavior;

[0081] Step S1243, in response to determining based on the lane matching result that the historical planned lane information is inconsistent with the current detected lane information, the current detected lane information in the initial lane model is corrected to obtain a target lane model.

[0082] The above-mentioned historical planned lane information refers to the specific information of the vehicle target lane list and reference lane list determined by the decision planning module in the previous data frame or processing cycle in the vehicle driving control system. Specifically, the historical planned lane information includes the stored reference lane list of the previous frame and the target lane list determined by the decision module in the previous frame, wherein the reference lane list of the previous frame represents the lane sequence that the vehicle is expected to follow according to the driving intention, and the target lane list determined in the previous frame represents the specific lane ID that the vehicle plans to change to.

[0083] The above-mentioned currently detected lane information refers to the specific information of the lane in which the vehicle is located, determined by the environment perception module in the vehicle driving control system based on the latest collected data in the current data frame. Specifically, the currently detected lane information includes but is not limited to the lane ID corresponding to the current position of the vehicle, and the geometric attributes of the lane, such as width, length and shape.

[0084] The above-mentioned lateral behavior refers to any lateral movement or change of direction of the vehicle relative to its driving direction during driving, including but not limited to changing lanes, avoiding obstacles, turning left, turning right, and lateral adjustments due to changes in road conditions.

[0085] Specifically, in the process of correcting the initial lane model, first, the reference lane list and target lane list corresponding to the historical data frame and the current lane corresponding to the current data frame are obtained based on the initial lane model. Further, the above information is matched to determine whether the target vehicle has lateral behaviors such as lane changing, obstacle avoidance and steering. Finally, if the target vehicle has lateral behaviors, the current detected lane information in the initial lane model needs to be corrected to obtain the target lane model.

[0086] For example, assuming that the target vehicle is initially located in lane ID2, in the constructed initial lane model, the reference lane list includes lanes ID1, ID2, and ID3, and the target lane list is lane ID2. However, the current detected lane information shows that the target vehicle has actually moved to lane ID1. At this time, the historical planned lane information and the current detected lane information do not match, indicating that the target vehicle has lateral behavior. It is necessary to correct the current detected lane information in the initial lane model from lane ID2 to lane ID1 to obtain the target lane model.

[0087] Based on the above steps S1241 to S1243, by real-time analysis of the differences between the historical planned lane information and the current detected lane information, the lateral behavior deviations that may occur in the target vehicle during driving can be timely identified and corrected, ensuring that the lane model matches the actual driving state of the vehicle, thereby improving the accuracy and real-time performance of the vehicle driving control system environmental model.

[0088] Figure 4 is a schematic diagram of another vehicle driving control method provided by an embodiment of the present application, such as Figure 4 As shown, when the vehicle is avoiding obstacles, it will drive to lane A, but the target lane of the vehicle at this moment should be lane B. At this time, lane correction is required to correct the target lane from lane A to lane B.

[0089] Optionally, step S1243, correcting the current detected lane information in the initial lane model to obtain the target lane model includes:

[0090] Step S21, obtaining the lateral distance between the target vehicle and the target lane;

[0091] Step S22, in response to the lateral distance being less than a preset distance, correcting the current lane in the initial lane model to a target lane to obtain a target lane model, wherein the preset distance is determined according to a lane width of the target lane.

[0092] The above lateral distance refers to the lateral offset of the target vehicle relative to the centerline of the target lane. Generally speaking, the lateral distance is determined by measuring or calculating the vertical distance between the position of the target vehicle and the centerline of the target lane, which is crucial for monitoring and controlling the position of the vehicle during lateral movement such as lane changing and obstacle avoidance.

[0093] The above-mentioned preset distance refers to a threshold standard used to determine whether the target vehicle has successfully transitioned to the target lane or is sufficiently close to the target lane. The preset distance is usually set based on the width of the target lane and a certain safety redundancy to ensure that when the lateral offset of the target vehicle is less than the preset distance, the vehicle is already in or close to the target lane, thereby avoiding unnecessary lane information correction or lane change operations. In the embodiment of the present application, half the lane width of the target lane is used as the preset distance.

[0094] For example, assuming that the width of the target lane is 3.5m, the preset distance is half the lane width of the target lane, i.e., 1.75m. If the lateral distance between the target vehicle and the center line of the target lane during the lane change process is measured to be 1.5m, which is less than the preset 1.75m, it means that the target vehicle has successfully entered the target lane, and the current lane in the initial lane model needs to be corrected to the target lane.

[0095] Based on the above steps S21 to S22, by real-time monitoring and calculating the lateral distance between the target vehicle and the center line of the target lane and comparing it with the preset distance, it is possible to timely identify whether the target vehicle has successfully transitioned to the target lane, ensuring that the vehicle driving control system can dynamically adjust the lane model according to the actual driving status of the vehicle, avoiding unnecessary lane information correction or lane changing operations caused by positioning errors or system noise, thereby improving the accuracy of the lane model and the real-time response of the system.

[0096] Optionally, in step S13, determining obstacle semantic information and road right relationship information based on the target lane model includes:

[0097] Step S131, obtaining obstacle path information and obstacle prediction information corresponding to the target obstacle based on the environmental perception information;

[0098] Step S132, projecting the target obstacle onto the target lane model based on the obstacle path information and the obstacle prediction information to obtain obstacle semantic information;

[0099] Step S133, determining the driving conflict type according to the obstacle semantic information, wherein the driving conflict type includes at least one of the following: lane keeping type, merging type, cutting-in type, and crossing conflict type;

[0100] Step S134: determining road right relationship information based on the driving conflict type.

[0101] The above obstacle path information refers to the real-time position, motion state and direction of obstacles obtained through multi-frame continuous detection and data fusion processing based on the current environmental perception data and historical environmental perception data. Specifically, the obstacle path information includes but is not limited to the dynamic and static properties of the obstacle, that is, whether it is moving and its acceleration state.

[0102] The above-mentioned obstacle prediction information refers to a set of information that combines the current map data and sensor perception information to estimate the movement trajectory, lane, and whether the obstacle has entered the intersection area in the future.

[0103] The above obstacle semantic information refers to the type of target obstacle, movement trend (such as lane keeping, lane changing, acceleration, deceleration, etc.), and lateral and longitudinal positions of the target obstacle relative to the target lane model after the target obstacle is projected onto the target lane model.

[0104] The above lane keeping type means that the target vehicle and an obstacle are in the same lane, and the obstacle has no intention or behavior to change lanes, that is, it keeps driving in the current lane.

[0105] The above merge type means that the target vehicle plans to or is changing to another lane, and there is an obstacle in the lane that is moving in the same or similar direction, but the obstacle is located in front of or behind the target vehicle and does not directly conflict with its lane change behavior.

[0106] The above-mentioned cut-in type means that the target vehicle is driving normally in a lane, while an obstacle in another lane intends to or is performing a lane change behavior, trying to cut into the lane where the target vehicle is located.

[0107] The above-mentioned intersection conflict type means that the target vehicle and the obstacle are located in different lanes, but their driving paths will intersect or converge at a certain point or area in the future.

[0108] Exemplarily, when there is an obstacle in the target lane model, first, based on the environmental perception information, multi-frame detection is used to obtain obstacle path information such as the lane in which the target obstacle is located and whether it is in an intersection; then, obstacle prediction information is obtained based on the prediction input, or based on the topological relationship in the cognitive module, or based on historical position fitting; finally, based on the obstacle path information and obstacle prediction information, the target obstacle is projected onto the target lane model to obtain obstacle semantic information.

[0109] Figure 5 is a schematic diagram of another vehicle driving control method provided by an embodiment of the present application, such as Figure 5 As shown in the figure, s is the lateral distance, and l is the vertical projection distance of the obstacle feature point in the local coordinate system. The s of obstacle feature point 1 is the s of the projection point, but the s of feature point 2 is the sum of the s of the projection point and the projection distance in the local coordinate system.

[0110] Exemplarily, after obtaining the obstacle path information and the obstacle prediction information, the obstacle path information and the obstacle prediction information are projected onto the target lane model based on the natural coordinate system, and the obstacle semantic information is calculated, and then the driving conflict type is determined based on the obstacle semantic information. Finally, the road right relationship information is determined based on the driving conflict type determined above.

[0111] Based on the above steps S131 to S134, by comprehensively analyzing the environmental perception information, obstacle path information, obstacle prediction information and the target lane model, real-time and detailed obstacle semantic information can be constructed, and then the type of driving conflict between the target vehicle and the obstacle can be accurately determined, ensuring that the vehicle driving control system can not only instantly identify obstacles in the surrounding environment, but also determine the road right relationship information according to the different driving conflict types of the obstacles, thereby providing a key basis for subsequent decision-making.

[0112] Optionally, in step S133, determining the driving conflict type according to the obstacle semantic information includes:

[0113] Step S31, obtaining an obstacle lane sequence and a reference line lane sequence;

[0114] Step S32, determining the obstacle current lane corresponding to the target obstacle according to the obstacle semantic information;

[0115] Step S33: in response to the overlap between the obstacle lane sequence and the reference line lane sequence, and the current lane of the obstacle belongs to the reference line lane sequence, determining that the driving conflict type is a lane keeping type.

[0116] The above obstacle lane sequence refers to the set of lanes that the obstacle may occupy on its driving path, based on the current lane of the obstacle, combined with its motion state and the prediction of the future behavior of the obstacle.

[0117] The above-mentioned reference line lane sequence refers to the lane sequence on the ideal driving path determined by the decision planning module based on the final driving intention of the target vehicle.

[0118] For example, it is assumed that the target obstacle is currently located in lane A, and according to its driving trend, it is expected that the target obstacle will continue to travel in lane A for a period of time in the future, that is, the obstacle lane sequence is lane A. At the same time, the reference line lane sequence of the target vehicle includes lane A and lane B. At this time, the obstacle lane sequence overlaps with the reference line lane sequence, and the current lane of the obstacle belongs to the reference line lane sequence, that is, the driving conflict type between the target vehicle and the obstacle is the lane keeping type.

[0119] Based on the above steps S31 to S33, when there is an overlap between the obstacle lane sequence and the reference lane sequence, and the current lane of the obstacle belongs to the reference lane sequence, it can be determined that the driving conflict type is the lane keeping type. By comparing the obstacle lane sequence of the target vehicle with the reference lane sequence in real time, the conflict type between the vehicle and the obstacle can be accurately identified and determined, thereby improving the overall driving safety and efficiency.

[0120] Optionally, the vehicle driving control method further includes: in response to the absence of overlap between the obstacle lane sequence and the reference line lane sequence, and the target vehicle and the target obstacle satisfying preset position conditions, determining that the driving conflict type is an intersection conflict type.

[0121] The above-mentioned preset position condition refers to the situation that although the obstacle and the target vehicle are not currently in the same lane sequence, based on the position, speed, acceleration and driving path prediction of the two, it is expected that they will intersect or be close to intersecting at a certain time point or section in the future.

[0122] For example, if the obstacle lane sequence has no intersection with the reference line lane sequence, but the target vehicle is in the intersection area and the obstacle has an intersection with the reference line lane sequence, it belongs to the intersection conflict type.

[0123] Based on the above optional embodiments, by real-time analysis and prediction of the possibility of future intersection between the target vehicle and the obstacle, it is possible to accurately determine whether the driving conflict type between the obstacle lane sequence and the reference line lane sequence is an intersection conflict type under the condition that the obstacle lane sequence and the reference line lane sequence do not overlap, so that the vehicle driving control system can identify and handle potential intersection conflicts in advance in complex road environments, such as intersections, multi-lane intersections and other scenarios, so that more timely and effective avoidance or adjustment measures can be taken to avoid collision risks and ensure driving safety and road traffic efficiency.

[0124] Optionally, in step S133, determining the driving conflict type according to the obstacle semantic information includes:

[0125] Step S41, obtaining an obstacle lane sequence, a reference line lane sequence and a driving behavior prediction result;

[0126] Step S42, determining the obstacle current lane corresponding to the target obstacle according to the obstacle semantic information;

[0127] Step S43, in response to the overlap between the obstacle lane sequence and the reference line lane sequence, and the current lane of the obstacle does not belong to the reference line lane sequence, determining the driving conflict type based on the overlap lane type and the driving behavior prediction result.

[0128] The above-mentioned overlapping lane type refers to the lanes that intersect or partially overlap in the obstacle lane sequence and the reference line lane sequence of the target vehicle. Specifically, when the lane sequence in which the target vehicle is traveling or plans to travel has at least one lane in common with the lane sequence on the ideal driving path determined by the decision planning module or has a spatially overlapping area, the above-mentioned lane is called an overlapping lane. The overlapping lane type can be divided into forward lanes and backward lanes according to their positional relationship in the reference line lane sequence. Among them, the forward lane refers to the lane in the obstacle lane sequence of the target vehicle that overlaps with the reference line lane sequence and is located within a certain distance in front of the target vehicle; the backward lane refers to the lane in the obstacle lane sequence of the target vehicle that overlaps with the reference line lane sequence and is located behind the target vehicle.

[0129] The above-mentioned driving behavior prediction results refer to the driving behavior and dynamics of the target vehicle in the future predicted by the vehicle driving control system based on the current state of the target vehicle, environmental conditions, road information and driving intention of the target vehicle, using machine learning, deep learning, mathematical model analysis, etc. Specifically, the driving behavior prediction results include but are not limited to predicting whether the target vehicle will accelerate, decelerate, change lanes, overtake, stop, pass through intersections or perform specific lane keeping strategies, as well as detailed information such as the time point, duration, speed change, acceleration change and specific driving path of these operations.

[0130] For example, when there is overlap between the obstacle lane sequence and the reference line lane sequence, and the current lane of the obstacle does not belong to the reference line lane sequence, it is necessary to determine whether the overlapping lane type belongs to the forward lane or the backward lane, and determine the driving conflict type in combination with the driving behavior prediction results.

[0131] Based on the above steps S41 to S43, by real-time analysis of the relative position of the target vehicle and the obstacle, the driving behavior prediction results of the target vehicle, and the overlapping lane type between the reference line lane sequence and the obstacle lane sequence, the driving conflict type can be more accurately identified and determined, thereby providing key decision-making basis for the vehicle driving control system and ensuring the safe driving of the vehicle under complex road conditions.

[0132] Optionally, in step S43, determining the driving conflict type based on the overlapping lane type and the driving behavior prediction result includes:

[0133] Step S431, if it is determined based on the driving behavior prediction result that the target vehicle performs a lane change behavior, the driving conflict type is determined to be a cut-in type;

[0134] Step S432, if it is determined based on the driving behavior prediction result that the target vehicle performs a tracking behavior, and the overlapping lane type is a forward lane, the driving conflict type is determined to be a merging type;

[0135] Step S4333: If it is determined based on the driving behavior prediction result that the target vehicle performs tracking behavior, and the overlapping lane type is a rear lane, the driving conflict type is determined to be a lane keeping type.

[0136] The above-mentioned tracking behavior refers to the behavior of the target vehicle driving along the center line of the current lane or its subsequent lane sequence without actively changing lanes or turning.

[0137] Specifically, when the target vehicle performs lane changing behavior, its driving conflict type is the cutting-in type; when the target vehicle performs tracking behavior and the overlapping lane type is the forward lane, its driving conflict type is the merging type; when the target vehicle performs tracking behavior and the overlapping lane type is the backward lane, its driving conflict type is the lane keeping type.

[0138] Based on the above steps S431 to S433, by real-time analysis of the predicted driving behavior results of the target vehicle and the relative position of the obstacle, combined with the overlap type of the lane sequence planned by the vehicle and the lane where the obstacle is located, the vehicle driving control system can accurately identify and classify the types of driving conflicts, including the "cut-in type" conflict caused by lane changing behavior, the "merge type" conflict caused by the front obstacle in the tracking behavior, and the "lane keeping type" conflict when the rear obstacle approaches. It can not only ensure the safe driving of the target vehicle under complex road conditions, but also provide key decision-making basis for the vehicle driving control system to adjust the vehicle's driving strategy, such as timely lane changing, deceleration or acceleration, to avoid collision risks and ensure driving safety.

[0139] Optionally, in step S133, determining the road right relationship information based on the driving conflict type includes:

[0140] Step S51, obtaining the obstacle type of the target obstacle;

[0141] Step S52: determining the right-of-way relationship information based on the driving conflict type, obstacle type and preset right-of-way rules.

[0142] The obstacle types of the above-mentioned target obstacles may include vehicles, pedestrians and non-motor vehicles.

[0143] For example, assuming that the target obstacle is a vehicle, when the driving conflict type is the lane keeping type, if the target obstacle is not behind the current reference line of the target vehicle, the target vehicle has the right of way; when the driving conflict type is the cut-in type, the cut-in vehicle does not have the right of way, that is, the target vehicle cannot forcibly cut into the lane of other vehicles when changing lanes; when the driving conflict type is the merging type or the crossing conflict type, it is necessary to further judge the turning attributes of the target obstacle and the target vehicle at the intersection, and compare the priorities of the two in the order of straight going, left turn, right turn and U-turn. If the turning attribute priority of the obstacle is higher than that of the target vehicle, the target obstacle has the right of way.

[0144] For example, if the target obstacle is a pedestrian or non-motor vehicle, if the reference line of the target vehicle intersects with the path of the pedestrian or non-motor vehicle, the driving conflict type is the intersection conflict type. In certain cases, if the pedestrian or non-motor vehicle is continuously within the reference line boundary of the target vehicle for a certain period of time, the driving conflict type is the lane keeping type. In the above two cases, pedestrians and non-motor vehicles are always given the right of way to give priority to the safety of pedestrians and non-motor vehicles.

[0145] Exemplarily, after determining the map and obstacle information, it is also necessary to improve the semantic information such as the passability, driving speed, safety, and meeting status of the reference line as an important reference for intention decision-making and reference line decision-making. First, it is necessary to perform collision detection on the reference line. If the minimum distance between the static obstacle and the reference line is less than the sum of half the vehicle width and the safety threshold, it means that the reference line is currently impassable. Secondly, the driving speed of the reference line adopts the speed of the nearest obstacle in front. If there is no obstacle in front, the driving speed of the reference line is set to the map speed limit of the reference line, and then compared with the preset threshold to determine the speed attribute of the reference line. Furthermore, safety mainly evaluates whether there is a conflict risk when the reference line is used as a target to perform lane change. If there are obstacles with similar speeds within a certain distance before and after the target vehicle or there are obstacles with a short collision time with the target vehicle, safety is not met. Finally, the risk of the oncoming road is classified based on the meeting status. If the oncoming vehicle enters the boundary of the target vehicle reference line, it belongs to the speed limit type; if the closest distance between the oncoming vehicle and the target vehicle reference line is less than the sum of half the vehicle width and the safety threshold, it belongs to the avoidance type; if the closest distance between the oncoming vehicle and the right boundary of the target vehicle reference line is less than the sum of the vehicle width and the safety threshold, it belongs to the parking type.

[0146] Based on the above steps S51 to S52, by real-time monitoring of the type of target obstacle and the type of driving conflict between it and the target vehicle, combined with preset right-of-way rules, the vehicle driving control system can dynamically determine the right-of-way relationship information to ensure that the intelligent driving vehicle can pass safely in various traffic scenarios.

[0147] Embodiment 2

[0148] The present application also provides a vehicle driving control device 60, please refer to Figure 6 , Figure 6 : is a structural diagram of a vehicle cabin wireless charging device provided in an embodiment of the present application, comprising:

[0149] Acquisition module 601, used to acquire environmental perception information, vehicle status information and decision planning information of the target vehicle;

[0150] A construction module 602 is used to construct a target lane model based on the environmental perception information, the vehicle state information and the decision planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment where the target vehicle is located;

[0151] A determination module 603 is used to determine obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle;

[0152] The control module 604 is used to control the target vehicle to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information.

[0153] Optionally, the construction module 602 is also used to: perform time synchronization processing on the environmental perception information, vehicle status information and decision planning information to obtain a time synchronization result; perform combination processing on the time synchronization result to obtain a lane list combination; perform conversion processing on the lane list combination to obtain an initial lane model; perform correction processing on the initial lane model to obtain a target lane model.

[0154] Optionally, the construction module 602 is also used to: obtain historical planned lane information and current detected lane information based on the initial lane model, wherein the historical planned lane information includes: a reference lane list and a target lane list corresponding to the historical data frame, the reference lane in the reference lane list is used to represent the decision driving lane at the historical decision moment, the target lane in the target lane list is used to represent the expected driving lane at the historical decision moment, and the current detected lane information includes: the current lane corresponding to the current data frame; use the historical planned lane information and the current detected lane information to perform matching processing to obtain a lane matching result, wherein the lane matching result is used to determine whether the target vehicle has lateral behavior; in response to determining that the historical planned lane information is inconsistent with the current detected lane information based on the lane matching result, the current detected lane information in the initial lane model is corrected to obtain a target lane model.

[0155] Optionally, the construction module 602 is also used to: obtain the lateral distance between the target vehicle and the target lane; in response to the lateral distance being less than a preset distance, correct the current lane in the initial lane model to the target lane to obtain a target lane model, wherein the preset distance is determined according to the lane width of the target lane.

[0156] Optionally, the determination module 603 is also used to: obtain obstacle path information and obstacle prediction information corresponding to the target obstacle based on the environmental perception information; project the target obstacle to the target lane model based on the obstacle path information and the obstacle prediction information to obtain obstacle semantic information; determine the driving conflict type based on the obstacle semantic information, wherein the driving conflict type includes at least one of the following: lane keeping type, merging type, cutting-in type, and intersection conflict type; determine the road right relationship information based on the driving conflict type.

[0157] Optionally, the determination module 603 is further used to: obtain an obstacle lane sequence and a reference line lane sequence; determine the obstacle current lane corresponding to the target obstacle based on the obstacle semantic information; in response to the overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane belongs to the reference line lane sequence, determine that the driving conflict type is a lane keeping type.

[0158] Optionally, the determination module 603 is further configured to: in response to the absence of overlap between the obstacle lane sequence and the reference line lane sequence, and the target vehicle and the target obstacle satisfying a preset position condition, determine that the driving conflict type is an intersection conflict type.

[0159] Optionally, the determination module 603 is also used to: obtain an obstacle lane sequence, a reference line lane sequence and a driving behavior prediction result; determine the obstacle current lane corresponding to the target obstacle based on the obstacle semantic information; in response to the overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane does not belong to the reference line lane sequence, determine the driving conflict type based on the overlapping lane type and the driving behavior prediction result.

[0160] Optionally, the determination module 603 is also used for: if it is determined based on the driving behavior prediction results that the target vehicle performs a lane changing behavior, determining the driving conflict type as a cutting-in type; if it is determined based on the driving behavior prediction results that the target vehicle performs a tracking behavior, and the overlapping lane type is a forward lane, determining the driving conflict type as a merging type; if it is determined based on the driving behavior prediction results that the target vehicle performs a tracking behavior, and the overlapping lane type is a backward lane, determining the driving conflict type as a lane keeping type.

[0161] Optionally, the determination module 603 is further used to: obtain the obstacle type of the target obstacle; and determine the right-of-way relationship information based on the driving conflict type, the obstacle type and the preset right-of-way rules.

[0162] The above-mentioned vehicle driving control device provided in the embodiment of the present application achieves the following technical effects: first, the environmental perception information, vehicle status information and decision-making planning information of the target vehicle are obtained, and a target lane model is constructed based on the above information; then, the obstacle semantic information and road right relationship information are determined based on the target lane model; finally, the target vehicle is controlled to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information, thereby achieving the purpose of improving the environmental model of the vehicle driving control system and enhancing the accuracy and comprehensiveness of the basis for driving decisions, thereby achieving the technical effect of improving the accuracy and stability of vehicle driving control, and then solving the technical problem of low accuracy and stability of the vehicle driving control method provided in the related art.

[0163] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0164] Embodiment 3

[0165] The present application also provides an electronic device 70, please refer to Figure 7 , Figure 7 It is a structural diagram of an electronic device provided in one embodiment of the present application, including a memory 710 and a processor 720, wherein the memory 710 is used to store computer programs; the processor 720 is used to execute the programs stored in the memory 710 to implement the vehicle driving control method introduced in any embodiment of the present application.

[0166] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0167] S1: Obtain the target vehicle’s environmental perception information, vehicle status information, and decision-making planning information;

[0168] S2: constructing a target lane model based on environmental perception information, vehicle state information and decision planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment where the target vehicle is located;

[0169] S3: Determine obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle;

[0170] S4: Control the target vehicle to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information.

[0171] The above-mentioned electronic device provided in the embodiment of the present application achieves the following technical effects: first, the environmental perception information, vehicle status information and decision-making planning information of the target vehicle are obtained, and a target lane model is constructed based on the above information; then, the obstacle semantic information and road right relationship information are determined based on the target lane model; finally, the target vehicle is controlled to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information, thereby achieving the purpose of improving the environmental model of the vehicle driving control system and enhancing the accuracy and comprehensiveness of the basis for driving decisions, thereby achieving the technical effect of improving the accuracy and stability of vehicle driving control, and then solving the technical problem of low accuracy and stability of the vehicle driving control method provided in the related technology.

[0172] It can be understood by those skilled in the art that Figure 7 The structure shown is for illustration only, and the electronic device may also be a terminal device such as a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, and a mobile Internet device (Mobile Internet Devices, MID for short). Figure 7 The structure of the electronic device is not limited. For example, the electronic device 70 may also include Figure 7More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 7 Different configurations shown.

[0173] Embodiment 4

[0174] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the vehicle driving control method introduced in any embodiment of the present application is implemented.

[0175] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0176] S1: Obtain the target vehicle’s environmental perception information, vehicle status information, and decision-making planning information;

[0177] S2: constructing a target lane model based on environmental perception information, vehicle state information and decision planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment where the target vehicle is located;

[0178] S3: Determine obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle;

[0179] S4: Control the target vehicle to make driving decisions based on the target lane model, obstacle semantic information and road right relationship information.

[0180] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0181] The above-mentioned storage medium provided in the embodiment of the present application realizes the following technical effects: first, the environmental perception information, vehicle status information and decision-making planning information of the target vehicle are obtained, and a target lane model is constructed based on the above information; then, the obstacle semantic information and road right relationship information are determined based on the target lane model; finally, the target vehicle is controlled to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information, thereby achieving the purpose of improving the environmental model of the vehicle driving control system and enhancing the accuracy and comprehensiveness of the basis for driving decisions, thereby achieving the technical effect of improving the accuracy and stability of vehicle driving control, and then solving the technical problem of low accuracy and stability of the vehicle driving control method provided in the related technology.

[0182] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0183] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0184] In this application, a plurality refers to two or more than two.

[0185] In this application, unless otherwise clearly defined, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0186] The terms "first", "second", "third", "fourth", etc. (if any) in this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence.

[0187] The term "and / or" in this application is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this application generally indicates that the associated objects before and after are in an "or" relationship.

[0188] If not otherwise specified, all steps of the present application may be performed sequentially or randomly. For example, if a method includes steps A and B, it means that the method may include steps A and B performed sequentially, or may include steps B and A performed sequentially. For example, if a method may also include step C, it means that step C may be added to the method in any order, for example, the method may include steps A, B, and C, or may include steps A, C, and B, or may include steps C, A, and B, etc.

[0189] The above are only preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A vehicle driving control method, characterized in that: include: Obtain the target vehicle's environmental perception information, vehicle status information, and decision-making planning information; Building a target lane model based on the environmental perception information, the vehicle state information and the decision planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment of the target vehicle; Determining obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle; The target vehicle is controlled to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information.

2. The method according to claim 1, characterized in that Constructing the target lane model based on the environmental perception information, the vehicle state information and the decision planning information includes: Performing time synchronization processing on the environment perception information, the vehicle state information and the decision planning information to obtain a time synchronization result; Combining the time synchronization results to obtain a lane list combination; Converting the lane list combination to obtain an initial lane model; The initial lane model is corrected to obtain the target lane model.

3. The method according to claim 2, characterized in that Correcting the initial lane model to obtain the target lane model includes: Based on the initial lane model, historical planned lane information and current detected lane information are obtained, wherein the historical planned lane information includes: a reference lane list and a target lane list corresponding to a historical data frame, the reference lane in the reference lane list is used to represent the decision lane at the historical decision moment, the target lane in the target lane list is used to represent the expected lane at the historical decision moment, and the current detected lane information includes: the current lane corresponding to the current data frame; Performing matching processing using the historical planned lane information and the current detected lane information to obtain a lane matching result, wherein the lane matching result is used to determine whether the target vehicle has a lateral behavior; In response to determining that the historical planned lane information is inconsistent with the current detected lane information based on the lane matching result, the current detected lane information in the initial lane model is corrected to obtain the target lane model.

4. The method according to claim 3, characterized in that Correcting the currently detected lane information in the initial lane model to obtain the target lane model includes: Obtaining a lateral distance between the target vehicle and a target lane; In response to the lateral distance being less than a preset distance, the current lane in the initial lane model is corrected to the target lane to obtain the target lane model, wherein the preset distance is determined according to a lane width of the target lane.

5. The method according to claim 1, characterized in that Determining the obstacle semantic information and the road right relationship information based on the target lane model includes: Obtaining obstacle path information and obstacle prediction information corresponding to the target obstacle based on the environmental perception information; Projecting the target obstacle onto the target lane model based on the obstacle path information and the obstacle prediction information to obtain the obstacle semantic information; Determining a driving conflict type according to the obstacle semantic information, wherein the driving conflict type includes at least one of the following: a lane keeping type, a merging type, a cutting type, and a crossing conflict type; The right-of-way relationship information is determined based on the driving conflict type.

6. The method according to claim 5, characterized in that Determining the driving conflict type according to the obstacle semantic information includes: Obtain obstacle lane sequence and reference line lane sequence; Determining an obstacle current lane corresponding to the target obstacle according to the obstacle semantic information; In response to the overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane belongs to the reference line lane sequence, the driving conflict type is determined to be a lane keeping type.

7. The method according to claim 6, characterized in that The method further comprises: In response to the absence of overlap between the obstacle lane sequence and the reference line lane sequence, and the target vehicle and the target obstacle satisfying a preset position condition, it is determined that the driving conflict type is an intersection conflict type.

8. The method according to claim 5, characterized in that Determining the driving conflict type according to the obstacle semantic information includes: Obtain obstacle lane sequence, reference line lane sequence and driving behavior prediction results; Determining an obstacle current lane corresponding to the target obstacle according to the obstacle semantic information; In response to overlap between the obstacle lane sequence and the reference line lane sequence, and the obstacle current lane does not belong to the reference line lane sequence, the driving conflict type is determined based on the overlap lane type and the driving behavior prediction result.

9. The method according to claim 8, characterized in that Determining the driving conflict type based on the overlapping lane type and the driving behavior prediction result includes: If it is determined based on the driving behavior prediction result that the target vehicle performs a lane change behavior, determining that the driving conflict type is a cut-in type; If it is determined based on the driving behavior prediction result that the target vehicle performs a tracking behavior, and the overlapping lane type is a forward lane, determining that the driving conflict type is a merging type; If it is determined based on the driving behavior prediction result that the target vehicle performs a tracking behavior, and the overlapping lane type is a rearward lane, the driving conflict type is determined to be a lane keeping type.

10. The method according to claim 5, characterized in that Determining the road right relationship information based on the driving conflict type includes: Obtaining the obstacle type of the target obstacle; The right-of-way relationship information is determined based on the driving conflict type, the obstacle type and a preset right-of-way rule.

11. A vehicle driving control device, characterized in that: include: An acquisition module is used to acquire the target vehicle's environmental perception information, vehicle status information, and decision-making planning information; A construction module, used to construct a target lane model based on the environmental perception information, the vehicle state information and the decision planning information, wherein the target lane model is used to describe the road structure and environmental characteristics in the driving environment of the target vehicle; A determination module, configured to determine obstacle semantic information and road right relationship information based on the target lane model, wherein the obstacle semantic information is used to characterize the motion state of the target obstacle, and the road right relationship information is used to characterize the driving conflict between the target vehicle and the target obstacle; A control module is used to control the target vehicle to make driving decisions based on the target lane model, the obstacle semantic information and the road right relationship information.

12. An electronic device, characterized in that: comprising a processor and a memory, wherein: Memory, used to store computer programs; A processor, used to execute a program stored in a memory to implement the method described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images