Method, device and equipment for generating escape trajectory of autonomous vehicle and storage medium

Abstract

The application discloses a method and device for generating an escape trajectory of an autonomous vehicle, an equipment and a storage medium, relates to the technical field of autonomous driving, and discloses a method for generating an escape trajectory of an autonomous vehicle, which comprises the following steps: when detecting that an autonomous light truck is trapped, acquiring vehicle trapping data, target region map data, road topology data, whole vehicle performance parameters and global road condition data; generating an initial escape trajectory according to the vehicle trapping data, the target region map data, the road topology data, the whole vehicle performance parameters and the global road condition data; checking and judging the safety of the initial escape trajectory to obtain a trajectory judgment result; and generating a target escape trajectory according to the trajectory judgment result. By applying the technical means of the present application, the escape trajectory can be generated and checked in combination with the vehicle trapping state, the road environment, the whole vehicle performance and the global road condition, so that the safety, feasibility and working condition adaptability of the escape trajectory of the autonomous vehicle are improved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a method, apparatus, device, and storage medium for generating traction trajectories for autonomous vehicles. Background Technology

[0002] As the application of autonomous driving technology in the logistics and transportation sector continues to deepen, Level 4 autonomous light trucks are gradually being applied to structured road scenarios such as park roads, port roads, and urban delivery roads. Structured roads typically feature clearly defined lane boundaries, clear traffic signs, and relatively fixed driving rules. However, in actual operation, vehicles may still become stuck due to road construction, temporary obstacles, changes in road surface adhesion conditions, limited road space, abnormal vehicle posture, or insufficient perception of the local environment. When a vehicle becomes stuck, it may be unable to continue along its original path, requiring a re-determination of the feasible driving area, vehicle direction, and control methods such as steering, acceleration, and braking. Because light trucks differ from ordinary passenger cars in terms of body size, wheelbase, minimum turning radius, load conditions, and power performance, the process of getting out of trouble also requires consideration of the vehicle's traffic capacity, the distribution of surrounding obstacles, road boundary restrictions, and the vehicle's current operating status to avoid collisions with obstacles or further deviation from the passable area.

[0003] Given the aforementioned technological background, traditional solutions typically rely on the vehicle's autonomous driving system to autonomously identify obstacles, determine drivable areas, and plan local paths based on local environmental information collected by onboard sensing devices such as LiDAR, cameras, and millimeter-wave radar. The system then controls the vehicle to move forward, backward, steer, or brake based on the planned path. When the onboard autonomous driving system is unable to extricate itself from a difficult situation, a remote operator can assess the situation using onboard video and vehicle status information, and control the vehicle to escape by remote takeover, remote driving, or issuing simple driving commands. However, traditional solutions primarily depend on limited local sensing information from the vehicle or real-time control by remote operators, making it difficult to accurately determine a safe escape control strategy that matches the vehicle's performance and the surrounding environment in complex and challenging scenarios. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, device, and storage medium for generating an escape trajectory for an autonomous vehicle, aiming to solve the technical problem that existing L4-level autonomous light trucks have low efficiency in determining escape control strategies when stuck on structured roads, and that these strategies are difficult to match with the vehicle's current state and the on-site environmental conditions.

[0005] To achieve the above objectives, this application proposes a method for generating an escape trajectory for an autonomous vehicle, the method comprising: When an autonomous light truck is detected to be stuck, the system acquires vehicle entrapment data, target area map data, road topology data, vehicle performance parameters, and global road condition data. An initial escape trajectory is generated based on the vehicle entanglement data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data; The initial escape trajectory is verified and safety is assessed to obtain the trajectory assessment result; Based on the trajectory analysis results, a target escape trajectory is generated.

[0006] In one embodiment, the step of verifying and assessing the initial escape trajectory to obtain the trajectory assessment result includes: The initial escape trajectory is pre-verified to obtain the pre-verification result, and the verification escape trajectory is determined based on the pre-verification result; Based on the vehicle's current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters, a safety assessment is performed on the verified escape trajectory to obtain the trajectory assessment result.

[0007] In one embodiment, the step of pre-verifying the initial escape trajectory, obtaining a pre-verification result, and determining the verified escape trajectory based on the pre-verification result includes: The initial escape trajectory is analyzed to obtain multiple trajectory segments arranged in the order of trajectory execution; Based on the trajectory segment, the target area map data, and the road topology data, the feasibility of the trajectory is verified to obtain the feasibility verification result. Based on the trajectory segment and the global road condition data, an obstacle collision risk verification is performed to obtain the collision risk verification result. Based on the trajectory segment and the vehicle performance parameters, the vehicle parameter matching degree is verified to obtain the vehicle parameter matching verification result. The pre-verification result is generated based on the passage feasibility verification result, the collision risk verification result, and the vehicle parameter matching verification result. When the pre-verification result is "pre-verification passed", the verification escape trajectory is determined based on the vehicle parameter matching verification result.

[0008] In one embodiment, the step of performing a safety assessment of the verified escape trajectory based on the vehicle's current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters to obtain the trajectory assessment result includes: The current environmental perception data is fused to obtain obstacle distribution data and passable area data; The operating status of the vehicle powertrain system, the vehicle braking system, and the vehicle steering system are determined based on the vehicle operating status data. Based on the vehicle traffic performance parameters and the verification escape trajectory, determine the trajectory curvature, turning space and passage distance corresponding to each trajectory segment; Based on the obstacle distribution data, the passable area data, the vehicle power system operating status, the vehicle braking system operating status, the vehicle steering system operating status, the trajectory curvature, the turning space, and the passage distance, environmental risk data, vehicle operating condition matching data, and trajectory adaptation data are determined. A comprehensive safety assessment is conducted based on the environmental risk data, the vehicle operating condition matching data, and the trajectory adaptation data to obtain the trajectory assessment result.

[0009] In one embodiment, the step of generating an initial escape trajectory based on the vehicle entrapment data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data includes: The vehicle entrapment data is analyzed to obtain the current position, current vehicle posture, current chassis condition, and current power status of the autonomous driving light truck. Candidate traffic areas and candidate trajectory nodes within the candidate traffic areas are determined based on the current location, the target area map data, the road topology data, and the global traffic data. Based on the vehicle performance parameters, the current vehicle posture, the current chassis condition, and the current power state, determine the trajectory curvature constraint, turning radius constraint, driving speed constraint, and braking distance constraint; A constrained escape trajectory is generated based on the candidate trajectory nodes, the trajectory curvature constraint, the turning radius constraint, the driving speed constraint, and the braking distance constraint; The initial escape trajectory is determined based on the trajectory length, trajectory continuity, road occupancy, and obstacle distribution of each constrained escape trajectory.

[0010] In one embodiment, the step of generating a target escape trajectory based on the trajectory analysis result includes: By analyzing the trajectory analysis results, the trajectory safety status, trajectory risk type, and risk trajectory segment can be obtained; When the trajectory safety status is that the verified escape trajectory meets the preset safety conditions, the verified escape trajectory is determined as the target escape trajectory; When the trajectory safety status is that the verified escape trajectory does not meet the preset safety conditions, trajectory replanning constraints are generated based on the trajectory risk type, the risk trajectory segment, the vehicle entrapment data, and the global road condition data. Based on the trajectory replanning constraints, the trajectory points, driving direction, and driving speed corresponding to the risk trajectory segment are adjusted to obtain the replanned escape trajectory; The replanned escape trajectory is verified and its safety is assessed to obtain an updated trajectory assessment result. When the updated trajectory assessment result indicates that the replanned escape trajectory meets the preset safety conditions, the replanned escape trajectory is determined as the target escape trajectory.

[0011] In one embodiment, after the step of generating the target escape trajectory based on the trajectory analysis result, the method further includes: The execution trajectory segment and the corresponding target vehicle status are determined based on the target escape trajectory. Based on the execution trajectory segment and the target vehicle state, target driving parameters, target steering parameters, and target braking parameters are generated. The autonomous driving light truck is controlled to perform an escape operation based on the target driving parameters, target steering parameters, and target braking parameters. The vehicle position, body posture, driving speed, and surrounding environment change data of the autonomous driving light truck are also acquired. The escape execution status is determined based on the vehicle position, vehicle posture, driving speed, and changes in the surrounding environment. When the autonomous driving light truck deviates from its trajectory during the escape execution state, the unexecuted trajectory segment is corrected, and an updated target escape trajectory is generated.

[0012] In addition, to achieve the above objectives, this application also proposes an autonomous vehicle traction trajectory generation device, which includes: an traction detection module, used to acquire vehicle traction data, target area map data, road topology data, vehicle performance parameters and global road condition data when an autonomous light truck is detected to be stuck. The trajectory generation module is used to generate an initial escape trajectory based on the vehicle entrapment data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data. The trajectory verification module is used to verify and assess the safety of the initial escape trajectory, and obtain the trajectory assessment result. The trajectory generation module is also used to generate a target escape trajectory based on the trajectory analysis results.

[0013] In addition, to achieve the above objectives, this application also proposes an autonomous vehicle traction trajectory generation device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the autonomous vehicle traction trajectory generation method described above.

[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the above-described method for generating an autonomous vehicle's escape trajectory.

[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the autonomous vehicle traction trajectory generation method described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: When an autonomous driving light truck is detected to be stuck, this solution acquires vehicle entrapment data, target area map data, road topology data, vehicle performance parameters, and global road condition data. Based on this multi-source data, an initial escape trajectory is generated. This process combines the vehicle's current entrapment state, the target area's road structure, the vehicle's own traffic capacity, and the overall road conditions, avoiding reliance solely on local vehicle-side perception information to determine the escape path. The initial escape trajectory is further validated and safety assessed. Based on the trajectory assessment results, a target escape trajectory is generated. This ensures that the generated initial escape trajectory undergoes further screening and adjustment before being used as the basis for vehicle escape. This addresses the problem of low efficiency in determining escape control strategies for existing L4-level autonomous driving light trucks stuck on structured roads, and the difficulty in matching these strategies with the vehicle's current state and environmental conditions. Compared to existing solutions that primarily rely on local vehicle-side perception or remote, real-time control, this solution improves the efficiency of generating escape trajectories for autonomous vehicles and ensures a high degree of matching between the generated target escape trajectory and vehicle performance, entrapment state, and road environment, thus enhancing the feasibility and safety of the escape trajectory. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating an embodiment of the method for generating an escape trajectory for an autonomous vehicle according to this application; Figure 2 A flowchart illustrating Embodiment 2 of the method for generating an escape trajectory for an autonomous vehicle in this application; Figure 3 This is a schematic diagram of the module structure of the autonomous vehicle traction trajectory generation device according to an embodiment of this application; Figure 4 This is a schematic diagram of the device structure of the hardware operating environment involved in the method for generating the escape trajectory of an autonomous vehicle in this embodiment of the application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] The main solution of this application embodiment is as follows: when an autonomous driving light truck is detected to be stuck, vehicle stuck data, target area map data, road topology data, vehicle performance parameters, and global road condition data are acquired; an initial escape trajectory is generated based on the vehicle stuck data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data; the initial escape trajectory is verified and safety assessed to obtain a trajectory assessment result; and a target escape trajectory is generated based on the trajectory assessment result.

[0024] In this embodiment, for ease of description, the following description will focus on the device that generates the escape trajectory of an autonomous vehicle.

[0025] Because existing L4 autonomous driving light trucks face low efficiency in determining traction control strategies when stuck on structured roads, and struggle to match these strategies with the vehicle's current state and environmental conditions, this application provides a solution. Upon detecting a stuck autonomous light truck, it acquires vehicle entrapment data, target area map data, road topology data, vehicle performance parameters, and global road condition data. Based on this multi-source data, an initial traction trajectory is generated. This process combines the vehicle's current entrapment state, the target area's road structure, the vehicle's own traffic capacity, and overall road conditions, avoiding reliance solely on local vehicle-side perception information to determine the traction path. Furthermore, the initial traction trajectory... The escape trajectory is verified and safety assessed, and a target escape trajectory is generated based on the obtained trajectory assessment results. This allows the generated initial escape trajectory to undergo further screening and adjustment before being used as the basis for vehicle escape. This solves the problem that existing L4 autonomous driving light trucks have low efficiency in determining escape control strategies when stuck on structured roads and it is difficult to match them with the current state of the vehicle and the on-site environmental conditions. Compared with existing solutions that mainly rely on local perception on the vehicle or real-time control by remote personnel, this solution can improve the efficiency of generating escape trajectories for autonomous vehicles and make the generated target escape trajectory have a high degree of matching with vehicle performance, vehicle stuck state and road environment, thereby improving the feasibility and safety of the escape trajectory.

[0026] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, such as an autonomous vehicle's escape trajectory generation device. The following description uses an autonomous vehicle's escape trajectory generation device as an example to illustrate this embodiment and the subsequent embodiments.

[0027] Based on this, embodiments of this application provide a method for generating an escape trajectory for an autonomous vehicle, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for generating an escape trajectory for an autonomous vehicle according to this application.

[0028] In this embodiment, the method for generating the escape trajectory of the autonomous vehicle includes steps S10 to S40: Step S10: When it is detected that the autonomous driving light truck is stuck, acquire vehicle stuck data, target area map data, road topology data, vehicle performance parameters and global road condition data; In this embodiment, the autonomous vehicle's escape trajectory generation device can be deployed on a remote control platform and interact bidirectionally with the vehicle-side autonomous driving unit via a wireless communication link. The remote control platform receives vehicle entrapment data and, in conjunction with target area map data, road topology data, vehicle performance parameters, and global road condition data, completes escape trajectory planning, pre-verification, manual review, and trajectory distribution. The vehicle-side autonomous driving unit combines real-time on-site perception data and the vehicle's current operating status to perform a safety assessment of the received escape trajectory and decides whether to execute the escape trajectory based on the safety assessment results.

[0029] Furthermore, the remote control platform sends trajectory data or control instructions to the vehicle-mounted autonomous driving unit to assist the vehicle in getting out of trouble. The specific control processes for the vehicle's power, braking, and steering are still completed by the vehicle-mounted autonomous driving unit. Before executing the escape trajectory, the vehicle-mounted autonomous driving unit retains the right to judge the safety of the escape trajectory. If the escape trajectory does not match the on-site environment or the vehicle's current operating condition, it can refuse to execute the escape trajectory and report the corresponding risk information to the remote control platform.

[0030] It should be noted that autonomous light trucks are light-duty trucks with advanced autonomous driving capabilities. They can autonomously complete driving operations such as driving, steering, speed adjustment, and braking on structured roads without real-time driver intervention. Autonomous light trucks are typically equipped with multiple types of sensing devices and computing units, enabling them to autonomously perceive their surroundings and plan their driving routes.

[0031] Additionally, being stuck refers to a situation where a vehicle, due to obstacles, abnormal road conditions, or limited operating space, is unable to continue driving using conventional autonomous driving strategies and remains stationary or unable to exit normally. Being stuck affects vehicle traffic efficiency and driving safety, and is a prerequisite for triggering the escape and handling process.

[0032] In addition, vehicle entrapment data comprises multi-dimensional vehicle status and environmental data collected when a vehicle becomes stuck. This includes vehicle location information, vehicle posture data, chassis condition data, and surrounding environment data collected by onboard sensing devices. Vehicle entrapment data comprehensively reflects the vehicle's own state and surrounding environment at the time of entrapment, providing a fundamental basis for subsequent trajectory planning.

[0033] Additionally, the target area map data is high-precision map data of the area where the vehicle is stuck, containing detailed geographic information such as precise road coordinates, lane line positions, traffic sign locations, and the locations of roadside facilities. The target area map data provides a spatial benchmark for the stuck area, ensuring that the generated escape trajectory accurately matches the actual road space.

[0034] Additionally, road topology data describes the connectivity and structural attributes of the road network, including the number of lanes, lane orientation, intersection connections, and lane traffic rules. Road topology data clearly defines the road connectivity logic within a stuck area, ensuring that the generated escape trajectory conforms to the spatial structure requirements for road traffic.

[0035] In addition, the overall vehicle performance parameters are the inherent performance indicators of the corresponding model of autonomous driving light truck, including the vehicle's minimum turning radius, power output parameters, braking performance parameters, and vehicle dimensions. These overall vehicle performance parameters constrain the shape and dynamic parameters of the escape trajectory, ensuring that the trajectory matches the vehicle's actual travel capabilities.

[0036] Additionally, global traffic data provides comprehensive road traffic conditions within the area where the vehicle is stranded, including traffic flow, temporary traffic control information, and the traffic status of surrounding roads. Global traffic data reflects road conditions over a wider area, preventing the vehicle from escaping onto other road sections with obstacles.

[0037] Understandably, the system continuously monitors the operating status, displacement data, and changes in the surrounding environment of the autonomous light truck, comparing its real-time driving status with preset entrapment criteria. When the autonomous light truck's status meets these criteria, it is determined that the truck is entrapped, triggering the data acquisition process. This involves collecting entrapment data from the onboard perception unit, positioning unit, and chassis control unit, retrieving target area map data and road topology data from the map database, retrieving vehicle performance parameters for the corresponding model from the vehicle parameter database, and obtaining global road condition data from the road condition information platform, thus completing the collection and aggregation of all required data.

[0038] In one feasible implementation, upon detecting that an autonomous light truck has become stuck, the vehicle-mounted autonomous driving unit obtains vehicle positioning information from the positioning unit, vehicle posture data from the posture detection unit, chassis operating condition data from the chassis control unit, surrounding environment data from the onboard sensing device, and the vehicle's current power parameters. The vehicle-mounted autonomous driving unit encapsulates the vehicle positioning information, vehicle posture data, chassis operating condition data, surrounding environment data, and vehicle current power parameters to obtain vehicle stuck data, and uploads the vehicle stuck data to a remote management platform. The remote management platform determines the target area based on the vehicle positioning information, retrieves the target area map data, road topology data, and global road condition data corresponding to the target area, and retrieves the corresponding vehicle performance parameters based on the vehicle model of the autonomous light truck.

[0039] Step S20: Generate an initial escape trajectory based on the vehicle entanglement data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data; It should be noted that the initial escape trajectory is a preliminary vehicle escape path generated based on multi-source data, including a sequence of path coordinates, suggested driving speeds, suggested steering angles, and suggested gear shifts. This initial escape trajectory is a preliminary plan and has not yet undergone verification or review.

[0040] Understandably, the target area map data serves as the spatial benchmark, and road topology data is used to clarify the connectivity between passable lanes and roads within the stuck area. Based on vehicle stuck data, the vehicle's current stuck position and initial posture are determined, with the vehicle's current position as the trajectory starting point and the normally passable road area as the trajectory ending point. Dynamic parameters such as the turning radius and acceleration / deceleration amplitude of the trajectory are constrained by vehicle performance parameters to ensure the trajectory shape conforms to the vehicle's physical traffic capacity. Simultaneously, global road condition data is referenced to avoid congested or controlled road sections, and a continuous and complete initial escape trajectory is generated according to the shortest path planning logic.

[0041] In one feasible implementation, step S20 may include steps S21 to S25: Step S21: Analyze the vehicle entrapment data to obtain the current position, current vehicle posture, current chassis condition, and current power status of the autonomous driving light truck; It should be noted that the current location refers to the specific spatial location of the autonomous light truck when it becomes stuck. This typically includes latitude and longitude coordinates, elevation data, and its relative position within the lane. The current location accurately marks the spatial point where the vehicle is stuck and serves as the starting reference point for subsequent trajectory planning.

[0042] Additionally, the current vehicle attitude data refers to the spatial attitude data of the autonomous driving light truck when it is stuck, including information such as vehicle yaw angle, roll angle, pitch angle, and vehicle orientation. The current vehicle attitude reflects the vehicle's tilt and orientation when stuck, providing a basis for steering planning at the beginning of the trajectory.

[0043] Additionally, the current chassis condition data represents the operational status of various chassis systems when an autonomous light truck becomes stuck, including information such as steering system angle, braking system pressure, transmission gear position, and wheel speed. This current chassis condition reflects the state of the chassis actuators when the vehicle is stuck, providing support for assessing the feasibility of trajectory execution.

[0044] Additionally, the current power status refers to the output status data of the power system when an autonomous light truck is stuck, including information such as engine output torque, motor output power, and available power margin of the power system. The current power status reflects the level of traction power that the vehicle can currently provide, and it is used to constrain the acceleration and climbing ability of the escape trajectory.

[0045] Understandably, the process involves parsing and processing the vehicle entrapment data packet to extract positioning data, attitude sensing data, chassis bus data, and powertrain data. From the positioning data, the current location of the autonomous light truck is extracted; from the attitude sensing data, the current vehicle attitude is extracted; from the chassis bus data, the current chassis condition is extracted; and from the powertrain data, the current power state is extracted. All parameters are then parsed, categorized, and stored.

[0046] Step S22: Determine candidate traffic areas and candidate trajectory nodes within the candidate traffic areas based on the current location, the target area map data, the road topology data, and the global traffic data; It should be noted that the candidate passable areas are selected based on the road conditions at the stuck location, and are the passable road space range that allows autonomous light trucks to drive out of the stuck position. The candidate passable areas must meet conditions such as the road space size being larger than the vehicle's external dimensions and the absence of permanent traffic restrictions. The candidate passable areas are the spatial basis for generating candidate trajectories.

[0047] Additionally, candidate trajectory nodes are discrete path points selected within the candidate traffic area, containing attribute information such as point coordinates, the lane to which the point belongs, and the traffic priority of the point. Candidate trajectory nodes are the basic units for constructing escape trajectories, and are used to connect them to form a complete escape path.

[0048] Understandably, a spatial range with a preset radius is defined centered on the current location of the autonomous light truck, using map data of the target area. Within this spatial range, road topology data is used to filter out passable lanes that comply with traffic rules, forming a preliminary passable area. Global traffic data is used to exclude road sections with congestion, traffic control, or construction, and areas with passable conditions are selected from the preliminary passable area as candidate passable areas. Within the candidate passable areas, path points are evenly selected at preset intervals, and additional points are selected at key locations such as lane change points and intersection junctions, forming all candidate trajectory nodes.

[0049] In one feasible implementation, the remote control platform also acquires historical escape data, which includes the vehicle model, vehicle location, vehicle posture, road topology features, historical escape trajectory, and trajectory execution results corresponding to the historical entrapment tasks. Based on the vehicle model of the autonomous light truck, the current vehicle posture, and the road topology features corresponding to the target area, historical escape records matching the current entrapment scenario are selected from the historical escape data.

[0050] Understandably, historical trajectory segments that have been successfully executed are extracted from the historical escape records, and these historical trajectory segments are matched with the candidate traffic areas. When a historical trajectory segment is located within a candidate traffic area, and the traffic requirements corresponding to the historical trajectory segment meet the current vehicle performance parameters and global road condition data, the trajectory node corresponding to the historical trajectory segment is used as a candidate trajectory node, or the selection priority of the corresponding candidate trajectory node is increased. Historical escape data is only used to assist in generating candidate trajectories; the final generated initial escape trajectory still needs to be verified based on the current vehicle entrapment data and the current road environment.

[0051] Step S23: Determine the trajectory curvature constraint, turning radius constraint, driving speed constraint, and braking distance constraint based on the vehicle performance parameters, the current vehicle posture, the current chassis condition, and the current power state. It should be noted that trajectory curvature constraint is a parameter constraint that limits the curvature of the escape trajectory. It is used to specify the maximum curvature at any point on the trajectory to prevent the curvature from exceeding the vehicle's steering capability. The trajectory curvature constraint directly determines the bending amplitude of the trajectory and is the core geometric constraint that ensures the trajectory's feasibility.

[0052] Additionally, the turning radius constraint is a parameter constraint that limits the turning radius of the escape trajectory. It includes a minimum turning radius threshold and a maximum turning radius threshold, used to constrain the radius values ​​of turning segments in the trajectory. The turning radius constraint must match the vehicle's steering performance limits, and it can prevent the trajectory from having small-radius turns that the vehicle cannot complete.

[0053] Additionally, speed constraints are parameter constraints that limit the speed of vehicles on each segment of the escape trajectory. These include maximum and minimum speed thresholds, used to specify the speed range for different trajectory segments. Speed ​​constraints must match the vehicle's power output capability and road conditions to ensure driving stability during the escape process.

[0054] Additionally, braking distance constraints are parameter constraints that limit the length of the braking segment in the escape trajectory. They specify the shortest distance required for the vehicle to decelerate from its current speed to a target speed. Braking distance constraints must match the vehicle's braking performance with the road surface adhesion conditions, and they can prevent the safety risk of insufficient braking distance in the trajectory.

[0055] Understandably, this involves retrieving inherent performance limits from the vehicle's performance parameters, such as the minimum turning radius, maximum rated power output, and rated braking deceleration. Combined with the current vehicle posture, the available steering direction and maximum steering angle margin corresponding to the current vehicle orientation are determined. Based on the current chassis condition, the currently available steering travel of the steering system and the currently available braking travel of the braking system are determined. Combined with the current power state, the maximum power output and maximum braking capacity of the vehicle are determined. Based on the inherent limits of the vehicle's performance parameters and the actual usable boundaries of the vehicle's real-time state, allowable ranges for trajectory curvature are defined, forming trajectory curvature constraints. Allowable ranges for turning radius are defined, forming turning radius constraints. Allowable ranges for driving speed are defined, forming driving speed constraints. Minimum braking distance requirements are defined, forming braking distance constraints.

[0056] Step S24: Generate a constrained escape trajectory based on the candidate trajectory nodes, the trajectory curvature constraint, the turning radius constraint, the driving speed constraint, and the braking distance constraint; It should be noted that the constrained escape trajectory is an escape path that meets the constraint requirements, generated based on candidate trajectory nodes under various performance constraints. The constrained escape trajectory satisfies all geometric and dynamic constraints and serves as a candidate for the subsequent selection of the optimal trajectory.

[0057] Understandably, the starting node is the candidate trajectory node corresponding to the current location of the autonomous light truck, and the ending node is the candidate trajectory node at the exit of the candidate passage area. Multiple sets of concatenated paths are selected from all candidate trajectory nodes, and each set of paths is sequentially subjected to constraint verification. The verification checks whether the curvature value of the path conforms to the trajectory curvature constraint, whether the radius of the turning segment conforms to the turning radius constraint, whether the driving speed of each segment of the path conforms to the driving speed constraint, and whether the length of the deceleration segment conforms to the braking distance constraint. Paths that pass all constraint verifications are retained, and each path that passes verification corresponds to the generation of a constrained escape trajectory.

[0058] Step S25: Determine the initial escape trajectory based on the trajectory length, trajectory continuity, road occupancy, and obstacle distribution of each constrained escape trajectory.

[0059] It should be noted that trajectory length refers to the total path length of a single constrained escape trajectory from the starting point to the ending point, reflecting the total distance the vehicle needs to travel along the trajectory to escape the trapped area. Trajectory length directly affects the time required for the escape operation and is one of the core indicators for evaluating trajectory passage efficiency.

[0060] Additionally, trajectory continuity refers to the smoothness of the path connection of the constraint escaping trajectory, reflecting whether there are abrupt changes, bends, breaks, or discontinuities between different segments of the trajectory. The better the trajectory continuity, the smoother the steering and speed adjustment operations during vehicle movement. Trajectory continuity is an important indicator for ensuring the stability of escaping difficult situations.

[0061] Additionally, road occupancy refers to the number and range of lanes occupied by the escape trajectory during its operation, as well as its impact on surrounding lanes for normal traffic. Road occupancy reflects the impact of the escape trajectory on overall road traffic order and is a key indicator for evaluating the compliance of the trajectory.

[0062] Additionally, obstacle distribution refers to the number, location, and distance from obstacles along and around the constraint escape trajectory. Obstacle distribution directly affects the collision risk during trajectory execution and is a core basis for evaluating trajectory safety.

[0063] Furthermore, the initial escape trajectory is a preliminary escape route determined after multi-dimensional screening. It includes a complete sequence of route points and corresponding driving parameters, and is the object of processing in subsequent verification and analysis stages. The initial escape trajectory has basic feasibility and rationality, and it must be further verified and reviewed before it can be used to perform the escape operation.

[0064] Understandably, the process involves meticulously analyzing each constraint-based escape trajectory's performance across various metrics, comparing the total driving length of different trajectories, and prioritizing the shorter trajectory. It also involves comparing the path connectivity of different trajectories, prioritizing those with smooth, abrupt paths. Furthermore, it involves comparing the road occupancy of different trajectories, prioritizing those that occupy fewer lanes and have less impact on normal traffic. Finally, it involves comparing the distribution of obstacles along different trajectories, prioritizing those with more distant and fewer obstacles. Finally, based on the combined results from these multiple dimensions, the constraint-based escape trajectory with the best overall performance is selected as the initial escape trajectory.

[0065] Step S30: Verify and assess the safety of the initial escape trajectory to obtain the trajectory assessment result; It should be noted that verification refers to algorithmic validation of the initial escape trajectory from a technical feasibility perspective. This primarily verifies whether the trajectory conforms to vehicle performance parameters, whether there is a risk of collision with obstacles, and whether it meets road space requirements. Verification can objectively filter out trajectories with obvious defects from a technical standpoint, and it reduces the workload of subsequent manual review.

[0066] In addition, safety assessment refers to a comprehensive evaluation of the initial escape trajectory from a driving safety perspective. This assessment primarily evaluates potential environmental risks, operational condition compatibility risks, and traffic rule risks that may arise during trajectory execution. Safety assessment can comprehensively identify potential safety hazards along the trajectory and ensure driving safety during the escape process.

[0067] Furthermore, the trajectory analysis results are conclusive data derived from verifying and assessing the safety of the initial escape trajectory. They include information such as the feasibility of the trajectory, the type of risk, the location of the risk, and the risk level. The trajectory analysis results are used to determine whether the initial escape trajectory can be used directly, and also provide directional guidance for trajectory adjustment and optimization.

[0068] Understandably, the algorithm is first pre-verified by comparing the initial escape trajectory with the vehicle's performance parameters to verify whether the turning radius and acceleration / deceleration amplitude at each point on the trajectory are within the vehicle's performance limits. The initial escape trajectory is then compared with environmental perception data from the target area map data and vehicle entrapment data to detect any obstacles intruding along the trajectory and determine if there is a collision risk. Next, road topology data and global road condition data are combined to verify whether the trajectory complies with road traffic rules and regional road condition requirements. After completing all verification and analysis items, all verification results are summarized to form the trajectory analysis result.

[0069] Step S40: Generate the target escape trajectory based on the trajectory analysis results.

[0070] It should be noted that the target escape trajectory is a formal driving path that has been verified and assessed for feasibility before being used to execute the escape operation. The target escape trajectory includes precise path coordinates, driving speed sequence, steering control command timing, and braking control command timing, among other things.

[0071] Understandably, if the trajectory analysis results show that the initial escape trajectory meets all traffic and safety requirements, then the initial escape trajectory is directly designated as the target escape trajectory. If the trajectory analysis results show that the initial escape trajectory has correctable local risks, then the parameters of the trajectory segments corresponding to the risk points are adjusted, correcting the turning radius, driving speed, or path deviation, and then re-verified. If the trajectory analysis results show that the initial escape trajectory has uncorrectable major defects, then the relevant data is retrieved again to perform trajectory planning. This process continues until a trajectory that meets all traffic and safety requirements is generated, and then the corresponding trajectory is designated as the target escape trajectory.

[0072] In one feasible implementation, step S40 may include steps S41 to S45: Step S41: Analyze the trajectory analysis results to obtain the trajectory safety status, trajectory risk type, and risk trajectory segment; It should be noted that the trajectory safety status is status information used to indicate whether the verification escape trajectory meets the execution conditions. The trajectory safety status is determined based on the trajectory analysis results and is used to distinguish between two cases: the verification escape trajectory meets the preset safety conditions and the trajectory does not meet the preset safety conditions.

[0073] Additionally, the trajectory risk type indicates the category of safety risks present in the verified escape trajectory. The trajectory risk type can correspond to environmental risks, mismatch between the vehicle's current operating conditions and the trajectory, or incompatibility between the trajectory and the vehicle's traffic performance, providing a basis for subsequently determining the trajectory adjustment direction.

[0074] Furthermore, the risk trajectory segment is the portion of the verified escape trajectory that is determined to pose a safety risk. A risk trajectory segment can consist of one or more consecutive trajectory points, used to limit the range of trajectories that need to be replanned or adjusted subsequently, avoiding unnecessary changes to the portion of the trajectory where no risk exists.

[0075] It is understood that the environmental risk assessment content, vehicle condition assessment content, and trajectory adaptability assessment content recorded in the trajectory assessment results are analyzed to determine whether the overall verification escape trajectory meets the preset safety conditions, and the trajectory safety status is generated accordingly. When the verification escape trajectory has a safety risk, the trajectory risk type is determined according to the assessment dimension that generates the safety risk, and the corresponding trajectory point or continuous trajectory part is determined as the risk trajectory segment, thereby obtaining the trajectory safety status, the trajectory risk type, and the risk trajectory segment.

[0076] Step S42: When the trajectory safety status is that the verified escape trajectory meets the preset safety conditions, the verified escape trajectory is determined as the target escape trajectory; Understandably, when the trajectory safety status indicates that the verification trajectory meets the preset safety conditions, it means that after verification and safety assessment, the verification trajectory does not pose any safety risks to the vehicle in terms of road traffic, vehicle operating conditions, and trajectory adaptation. Therefore, the verification trajectory will not be replanned or partially adjusted; it will be directly determined as the target trajectory, so that the target trajectory can serve as the trajectory basis for subsequent escape operations.

[0077] Step S43: When the trajectory safety status is that the verified escape trajectory does not meet the preset safety conditions, generate trajectory replanning constraints based on the trajectory risk type, the risk trajectory segment, the vehicle entrapment data, and the global road condition data. It should be noted that trajectory replanning constraints are the restrictions that must be met when regenerating or adjusting the escape trajectory. Trajectory replanning constraints are used to limit the range of the trajectory that needs to be adjusted, the risk factors that need to be avoided, and the vehicle status and road traffic requirements that the adjusted trajectory should meet.

[0078] Understandably, when the trajectory safety status indicates that the verification of the escape trajectory does not meet the preset safety conditions, the first step is to determine the source of risk that prevents the verification of the escape trajectory from being executed safely, based on the trajectory risk type. Then, the trajectory range that needs to be adjusted is determined based on the risk trajectory segment. The limitations on trajectory adjustment caused by the vehicle's current position, body posture, chassis condition, and power status are determined by combining the vehicle's stuck data. The traffic conditions of the roads surrounding the risk trajectory segment and the range of roads available for replanning are determined by combining global road condition data. The above-mentioned sources of risk, adjustment range, vehicle status restrictions, and road traffic restrictions are summarized to generate trajectory replanning constraints.

[0079] In one feasible implementation, when the trajectory safety status indicates that the verified escape trajectory does not meet the preset safety conditions, the vehicle-mounted autonomous driving unit prohibits the generation of vehicle control parameters based on the verified escape trajectory and refuses to execute the verified escape trajectory. The vehicle-mounted autonomous driving unit encapsulates the trajectory analysis result, the trajectory risk type, the risk trajectory segment, and the current vehicle entrapment data, generates a trajectory replanning request, and sends the trajectory replanning request to the remote management and control platform.

[0080] Understandably, the remote control platform determines the source and location of the risk that prevents the verification escape trajectory from being executed based on the trajectory replanning request, and determines the range of passable roads around the risk trajectory segment in conjunction with the global road condition data, thereby generating the trajectory replanning constraint. The remote control platform adjusts or regenerates the escape trajectory based on the trajectory replanning constraint, and then re-issues the regenerated escape trajectory to the vehicle-side autonomous driving unit for safety assessment.

[0081] Step S44: Adjust the trajectory points, driving direction and driving speed corresponding to the risk trajectory segment according to the trajectory replanning constraints to obtain the replanned escape trajectory; It should be noted that trajectory points are path location data that constitute the escape trajectory. Multiple trajectory points are connected according to the vehicle's predetermined driving sequence to form a complete escape trajectory. Trajectory points are used to indicate the locations that the vehicle needs to pass through in sequence during the escape process and to determine the specific spatial range of the risk trajectory segment.

[0082] Additionally, the driving direction is the direction of movement of the vehicle as it passes through each point on the escape route. The driving direction is used to determine the vehicle's movement when passing through risky sections of the trajectory and is adapted to the vehicle's current position, body posture, and surrounding passable space.

[0083] Furthermore, driving speed refers to the speed information corresponding to the vehicle passing through each trajectory point or trajectory segment. The driving speed needs to match the vehicle's current power status, road conditions, and safety requirements of risky trajectory segments to avoid increasing the risk of collisions or trajectory deviations due to mismatched speed settings.

[0084] Furthermore, the replanned escape trajectory is the escape trajectory formed after replanning or adjusting the risk trajectory segment. The replanned escape trajectory retains the portion of the verified escape trajectory that does not pose a safety risk, and eliminates or reduces the safety risks identified in the trajectory assessment results by adjusting the risk trajectory segment.

[0085] Understandably, the process involves determining the trajectory points that need adjustment within the risky trajectory segment based on trajectory replanning constraints, and changing the position and connection relationships of the corresponding trajectory points within the passable area defined by the trajectory replanning constraints; determining the corresponding driving direction based on the vehicle's current attitude and the arrangement of the adjusted trajectory points, and adjusting the corresponding driving speed based on the vehicle's operating status and the road conditions of the risky trajectory segment; and connecting the adjusted trajectory points, driving direction, and driving speed with other trajectory sections without safety risks to obtain the replanned escape trajectory.

[0086] Step S45: Verify and assess the safety of the replanned escape trajectory to obtain an updated trajectory assessment result. When the updated trajectory assessment result indicates that the replanned escape trajectory meets the preset safety conditions, the replanned escape trajectory is determined as the target escape trajectory.

[0087] Understandably, after obtaining the replanned escape trajectory, the feasibility of the replanned escape trajectory, the vehicle's working condition adaptability, and the trajectory safety are reassessed using the same verification and safety assessment methods as the initial escape trajectory, resulting in an updated trajectory assessment result. When the updated trajectory assessment result indicates that the replanned escape trajectory meets the preset safety conditions, it means that the adjusted risk trajectory segment has met the safety requirements corresponding to the current vehicle status and the on-site road environment. Therefore, the replanned escape trajectory is determined as the target escape trajectory.

[0088] In one feasible implementation, after determining the target escape trajectory, the remote control platform encapsulates and encrypts the target escape trajectory data to obtain a target escape trajectory data packet, and sends the target escape trajectory data packet to the vehicle-side autonomous driving unit via a wireless communication link. The vehicle-side autonomous driving unit receives the target escape trajectory data packet, parses it to obtain the target escape trajectory, and determines multiple trajectory segments to be executed sequentially based on the arrangement of trajectory points in the target escape trajectory, providing a trajectory basis for subsequent segment-by-segment execution of the escape operation.

[0089] In this implementation, when generating the target escape trajectory data packet, the remote control platform configures a trajectory version identifier for the target escape trajectory and assigns a corresponding trajectory execution sequence number to each trajectory segment. After receiving the target escape trajectory data packet, the vehicle-side autonomous driving unit compares the trajectory version identifier with the trajectory version identifier corresponding to the current escape task and determines whether the trajectory execution sequence numbers corresponding to each trajectory segment are consecutive.

[0090] Furthermore, the vehicle-mounted autonomous driving unit determines, based on the current position and vehicle posture of the autonomous light truck, whether the first trajectory segment to be executed can be continuously connected with the current position and vehicle posture. If the trajectory version identifier matches the current escape task, the trajectory execution sequence number is consecutive, and the first trajectory segment to be executed can be continuously connected with the vehicle's current state, the target escape trajectory is determined to meet the execution conditions. If any verification fails, the execution of the target escape trajectory is suspended, and trajectory reception anomaly information is reported to the remote control platform.

[0091] In one possible implementation, steps S50 to S80 may be included after step S40: Step S50: Determine the execution trajectory segment and the target vehicle state corresponding to the execution trajectory segment based on the target escape trajectory; It should be noted that the execution trajectory segment is the portion of the trajectory that the autonomous driving light truck needs to execute from the target escape trajectory. The target escape trajectory can be divided into multiple continuous trajectory segments according to the driving sequence during the vehicle's escape process. The execution trajectory segment is used to clarify the road location and corresponding driving range that the autonomous driving light truck needs to pass through.

[0092] Additionally, the target vehicle state refers to the vehicle operating state that the autonomous light truck should achieve or maintain when executing the corresponding execution trajectory segment. The target vehicle state is used to characterize the driving direction, driving speed, vehicle posture, and steering and braking states of the autonomous light truck when passing through the execution trajectory segment, providing a basis for generating vehicle control parameters.

[0093] Understandably, based on the arrangement order of the trajectory segments in the target escape trajectory, and combined with the current position of the autonomous driving light truck in the target escape trajectory, the trajectory segments that have not yet been executed and are connected to the current position are determined, and the trajectory segments connected to the current position are determined as the executed trajectory segments; further, the vehicle operation requirements associated with the executed trajectory segments in the target escape trajectory are read, and the driving direction, driving speed and vehicle posture of the autonomous driving light truck when passing through the executed trajectory segment are determined, so as to obtain the target vehicle state corresponding to the executed trajectory segment.

[0094] Step S60: Based on the execution trajectory segment and the target vehicle state, generate target driving parameters, target steering parameters and target braking parameters, and control the autonomous driving light truck to perform an escape operation based on the target driving parameters, target steering parameters and target braking parameters, and obtain the vehicle position, body posture, driving speed and surrounding environment change data of the autonomous driving light truck; It should be noted that the target driving parameters are control parameters used to control the autonomous light truck to generate driving power that matches the state of the target vehicle. The target driving parameters are determined based on the driving direction and speed corresponding to the execution trajectory segment, and are used to control the autonomous light truck to move forward or backward along the execution trajectory segment.

[0095] Additionally, the target steering parameter is a control parameter used to control the autonomous light truck to change or maintain its driving direction. The target steering parameter is determined based on the extension direction of the execution trajectory segment, the trajectory changes, and the state of the target vehicle, and is used to control the autonomous light truck to complete the corresponding steering action along the execution trajectory segment.

[0096] Furthermore, the target braking parameter is a control parameter used to control the autonomous light truck to reduce its speed or stop. The target braking parameter is determined based on the driving speed corresponding to the target vehicle state and the traffic conditions of the execution trajectory segment, and is used to ensure that the autonomous light truck maintains a driving state that matches the execution trajectory segment during the extrication operation.

[0097] Additionally, the escaping-difficulty operation is a vehicle control process in which the autonomous light truck drives, steers, adjusts its speed, and brakes according to the execution trajectory segment to gradually leave the trapped area. The escaping-difficulty operation is autonomously executed by the autonomous light truck based on the target drive parameters, the target steering parameters, and the target braking parameters.

[0098] Furthermore, vehicle position is data used to indicate the location of the autonomous light truck within the escape area. Vehicle position reflects whether the autonomous light truck is moving along the execution trajectory segment and provides a positional basis for subsequent determination of whether the vehicle has deviated from its trajectory.

[0099] Additionally, vehicle posture data represents the vehicle's orientation and state during the autonomous light truck's obstacle avoidance operation. Vehicle posture is used to determine whether the autonomous light truck's actual orientation matches the target vehicle state corresponding to the execution trajectory segment.

[0100] Furthermore, the driving speed is data used to represent the actual speed at which the autonomous light truck moves during the extrication operation. The driving speed is used to determine whether the autonomous light truck passes through the execution trajectory segment at the speed corresponding to the target vehicle state.

[0101] Additionally, the surrounding environment change data indicates changes in the road environment surrounding the autonomous light truck during the extrication operation. This data reflects changes in the location of existing obstacles or the addition of new obstacles, providing a basis for determining whether the execution trajectory segment still meets the execution conditions.

[0102] It is understood that, based on the extension direction and trajectory change of the execution trajectory segment, as well as the driving direction and speed corresponding to the target vehicle state, the target driving parameters for controlling the vehicle's driving power, the target steering parameters for controlling the vehicle's driving direction, and the target braking parameters for controlling the vehicle's deceleration or stopping are determined respectively. The target driving parameters, the target steering parameters, and the target braking parameters are sent to the vehicle actuators corresponding to the autonomous driving light truck, so that the autonomous driving light truck completes driving, steering, speed adjustment, and braking according to the execution trajectory segment. During the autonomous driving light truck's escape operation, the actual position, actual vehicle posture, and actual driving speed of the autonomous driving light truck are continuously collected, and the changes in the surrounding environment of the vehicle are obtained through onboard sensing devices to obtain the vehicle position, the vehicle posture, the driving speed, and the surrounding environment change data.

[0103] In one feasible implementation, during the autonomous driving light truck's traction maneuver, the vehicle-mounted autonomous driving unit transmits data on the vehicle's location, vehicle posture, driving speed, and changes in the surrounding environment to a remote control platform according to a preset upload cycle. The remote control platform determines the actual driving trajectory of the autonomous driving light truck based on the continuously received vehicle location data, determines the traction progress based on the correspondence between the actual driving trajectory and the target traction trajectory, and monitors the traction process of the autonomous driving light truck in conjunction with the vehicle posture, driving speed, and changes in the surrounding environment data.

[0104] Step S70: Determine the extrication execution status based on the vehicle position, vehicle posture, driving speed, and surrounding environment change data; It should be noted that the "extrication execution status" is a state information used to indicate the actual situation of the autonomous light truck currently executing the target extrication trajectory. The extrication execution status can reflect whether the autonomous light truck is driving normally according to the execution trajectory segment, and whether the actual driving process of the autonomous light truck is affected by changes in the surrounding environment.

[0105] Understandably, the vehicle position is compared with the trajectory position corresponding to the execution trajectory segment to determine whether the actual driving position of the autonomous light truck is within the passage range corresponding to the execution trajectory segment; the vehicle posture and driving speed are compared with the vehicle posture and driving speed corresponding to the target vehicle state to determine whether the actual orientation and actual speed of the autonomous light truck meet the execution requirements of the execution trajectory segment; at the same time, based on the surrounding environment change data, it is determined whether there are environmental changes around the execution trajectory segment that affect the continued passage of the autonomous light truck; and the judgment results of the vehicle position, the vehicle posture, the driving speed, and the surrounding environment change data are combined to determine the escape execution state.

[0106] Step S80: When the escape execution state indicates that the autonomous driving light truck has deviated from its trajectory, the unexecuted trajectory segment is corrected to generate an updated target escape trajectory.

[0107] It should be noted that the unexecuted trajectory segment is the portion of the target escape trajectory that has not yet been completed by the autonomous driving light truck. The unexecuted trajectory segment is located after the current position of the autonomous driving light truck and may include the portion of the currently executed trajectory segment that has not yet been traversed, as well as other trajectory segments to be executed later.

[0108] Furthermore, the updated target escape trajectory is a new escape trajectory formed after correcting the unexecuted trajectory segment. The updated target escape trajectory is based on the actual vehicle position and body posture of the autonomous driving light truck after the trajectory deviation, and is used to reguide the autonomous driving light truck to complete the subsequent escape operation.

[0109] In one feasible implementation, after generating the updated target escape trajectory, the starting position and starting direction of the first unexecuted trajectory segment in the updated target escape trajectory are determined based on the current vehicle position and body posture of the autonomous driving light truck, and it is determined whether the starting position and the starting direction are continuously connected with the current vehicle position and body posture of the autonomous driving light truck.

[0110] Furthermore, when the first unexecuted trajectory segment continuously connects with the vehicle's current state, a safety assessment is performed on the updated target escape trajectory based on current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters. If the updated target escape trajectory meets preset safety conditions, the escape operation continues according to the updated target escape trajectory. If the updated target escape trajectory does not meet preset safety conditions, the autonomous driving light truck remains stationary, and a trajectory replanning request is sent to the remote control platform.

[0111] Understandably, when the autonomous driving light truck deviates from its trajectory in the "getting out of trouble" execution state, the actual execution progress of the autonomous driving light truck in the target getting out of trouble trajectory is determined based on the vehicle position, and the completed trajectory segments are distinguished from the unexecuted trajectory segments. The current vehicle position and vehicle posture are used as the starting state of the subsequent trajectory, and the trajectory position and driving direction that need to be adjusted in the unexecuted trajectory segments are determined in combination with the surrounding environment change data. The connection relationship between the unexecuted trajectory segments and the vehicle position is corrected. The corrected unexecuted trajectory segments are used as the trajectory content to be executed later, and the updated target getting out of trouble trajectory is generated.

[0112] In one feasible implementation, when the "getting out of trouble" execution state indicates that the autonomous driving light truck has deviated from its trajectory, the remote control platform generates a trajectory correction command based on the vehicle's position, vehicle posture, and the unexecuted trajectory segment, and sends the trajectory correction command to the vehicle-mounted autonomous driving unit. The vehicle-mounted autonomous driving unit corrects the unexecuted trajectory segment according to the trajectory correction command, and after completing a safety assessment of the corrected unexecuted trajectory segment, continues to execute the get-out-of-trouble operation.

[0113] Furthermore, when the surrounding environment change data indicates the appearance of a sudden obstacle within the target passage area, or when the vehicle operation status data indicates that the vehicle currently lacks the conditions to continue performing the extrication operation, the remote control platform generates an emergency stop command and sends it to the vehicle-mounted autonomous driving unit. The vehicle-mounted autonomous driving unit controls the autonomous light truck to stop driving according to the emergency stop command, and sends the vehicle's position, vehicle posture, and surrounding environment data after stopping to the remote control platform, so that the remote control platform can regenerate the extrication trajectory.

[0114] In this embodiment, when the autonomous driving light truck is detected to have escaped the stuck area, the autonomous driving light truck is controlled to exit the escape mode, resume normal autonomous driving state, and send an escape completion signal to the remote control platform.

[0115] It should be noted that the "getting out of trouble complete" signal is a status information indicating that the autonomous driving light truck has left the stuck area and completed the current get-out-of-trouble task. The "getting out of trouble complete" signal may include the task identifier corresponding to the get-out-of-trouble task, the get-out-of-trouble status, and the vehicle's position at the time of completion, providing a basis for the remote control platform to end the current get-out-of-trouble task.

[0116] Understandably, the system determines whether the autonomous driving light truck has left the trapped area based on its vehicle position, and whether it has entered a road area where it can drive normally based on its vehicle posture, speed, and changes in the surrounding environment. When the autonomous driving light truck has left the trapped area and meets the conditions for normal autonomous driving, the system stops calling the vehicle control parameters corresponding to the target escape trajectory, exits the escape mode, resumes normal autonomous driving, and simultaneously sends an escape completion signal to the remote control platform.

[0117] This embodiment provides a method for generating an escape trajectory for an autonomous vehicle. When an autonomous light truck is detected to be stuck, the method acquires vehicle entrapment data, target area map data, road topology data, vehicle performance parameters, and global road condition data. Based on the above data, an initial escape trajectory is generated, and the initial escape trajectory is verified and safety assessed. Finally, a target escape trajectory is generated based on the trajectory assessment results. This method combines the vehicle's entrapment state, road structure, vehicle traffic capacity, and overall road conditions to determine an executable escape trajectory, ensuring that the generated target escape trajectory matches the vehicle's current state and the surrounding environment.

[0118] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 Step S30 of the method for generating the escape trajectory of the autonomous vehicle includes steps S31 to S32: Step S31: Perform a pre-verification on the initial escape trajectory to obtain the pre-verification result, and determine the verification escape trajectory based on the pre-verification result; It should be noted that the pre-verification result is obtained after checking the feasibility of the initial escape trajectory, the risk of obstacle collision, and the matching degree of vehicle parameters. The pre-verification result is used to indicate whether the initial escape trajectory can be traversed within the target area, whether there is a risk of collision with obstacles, and whether it matches the overall vehicle performance parameters of the autonomous driving light truck.

[0119] Additionally, the verified escape trajectory is an escape trajectory that has undergone pre-verification and is eligible to enter the vehicle-side safety assessment process. The verified escape trajectory originates from the initial escape trajectory and is used to provide a trajectory basis for subsequent safety assessments based on vehicle on-site perception data and the vehicle's current operating status.

[0120] Understandably, the initial escape trajectory is first mapped to the target area map data and the road topology data to determine whether the initial escape trajectory is within the range of passable roads and whether it can continuously extend from the current position of the autonomous light truck to outside the trapped area, thereby completing the trajectory feasibility check. Further, combined with the global road condition data, it is determined whether there are obstacles affecting vehicle passage along the road sections traversed by the initial escape trajectory, thereby completing the obstacle collision risk check. Simultaneously, the steering and passage requirements corresponding to the initial escape trajectory are compared with the vehicle performance parameters to determine whether the autonomous light truck can complete the journey according to the initial escape trajectory, thereby completing the vehicle parameter matching check. The results of all checks are summarized to obtain the pre-verification result, and when the pre-verification result indicates that the initial escape trajectory has passed the pre-verification, the initial escape trajectory is determined as the verified escape trajectory.

[0121] In one feasible implementation, step S31 may include steps S311 to S316: Step S311: Analyze the initial escape trajectory to obtain multiple trajectory segments arranged in the order of trajectory execution; It should be noted that the trajectory execution order refers to the sequence in which the autonomous light truck traverses each part of the trajectory when executing the initial escape trajectory. The trajectory execution order is determined by the connection relationship between the trajectory positions in the initial escape trajectory and the vehicle's direction of travel, and is used to maintain the continuity between multiple trajectory segments.

[0122] Additionally, a trajectory segment is a portion of the initial escape trajectory that has a continuous positional relationship. Multiple trajectory segments are connected sequentially according to the trajectory execution order to jointly form the initial escape trajectory extending from the current position of the autonomous driving light truck to outside the trapped area.

[0123] It is understandable that the initial escape trajectory includes the trajectory position, trajectory connection relationship, and vehicle travel direction. Based on the continuity between adjacent trajectory positions, the start and end positions of each part of the trajectory are determined, and the continuous trajectory between adjacent start and end positions is divided into a trajectory segment. Following the travel direction of the autonomous driving light truck when it leaves the trapped area from its current position, the... Step S312: Perform trajectory feasibility verification based on the trajectory segment, the target area map data, and the road topology data to obtain the feasibility verification result. It should be noted that the trajectory feasibility verification is a process used to determine whether each trajectory segment is located within the passable road range of the target area, and whether adjacent trajectory segments can be passed continuously according to the trajectory execution order. The trajectory feasibility verification mainly relies on the road range and road connection relationship to determine whether the initial escape trajectory has basic passability conditions.

[0124] Additionally, the feasibility verification result is obtained after verifying the road location and connectivity of each trajectory segment. The feasibility verification result indicates whether each trajectory segment is located in a passable area and whether multiple trajectory segments can form a continuous escape route.

[0125] Understandably, the trajectory positions corresponding to each trajectory segment are correlated with the road ranges in the target area map data to determine whether each trajectory segment is located within the road boundaries of the target area and to identify whether each trajectory segment passes through impassable areas; further, the connection relationship between adjacent roads is determined based on the road topology data, and it is determined whether the end and start positions of adjacent trajectory segments are located on roads that can be continuously traveled; the road range determination results of each trajectory segment and the connection relationship determination results of adjacent trajectory segments are summarized to obtain the travel feasibility verification result.

[0126] Step S313: Perform obstacle collision risk verification based on the trajectory segment and the global road condition data to obtain the collision risk verification result; It should be noted that the obstacle collision risk verification is a process used to determine whether there is a risk of collision with obstacles in the road when the autonomous light truck travels along each of the aforementioned trajectory segments. The obstacle collision risk verification is completed by comparing the road positions traversed by the trajectory segments with the road occupancy status reflected in the global road condition data.

[0127] Additionally, the collision risk verification result is obtained by assessing the collision risk of obstacles corresponding to each of the aforementioned trajectory segments. The collision risk verification result indicates whether the initial escape trajectory passes through road locations with obstacles or other obstacles affecting vehicle passage.

[0128] It is understood that, based on each trajectory segment, the road locations that the autonomous light truck needs to pass through when executing the initial escape trajectory are determined, and the obstacle distribution and road occupancy status of the corresponding road locations are obtained from the global road condition data; the road locations passed through by each trajectory segment are matched with the locations of obstacles to determine whether the obstacles are within the vehicle's passage range corresponding to the trajectory segment; based on the judgment results corresponding to each trajectory segment, it is determined whether there is an obstacle collision risk for each trajectory segment, and the collision risk verification results are summarized.

[0129] Step S314: Perform vehicle parameter matching degree verification based on the trajectory segment and the vehicle performance parameters to obtain vehicle parameter matching verification result; It should be noted that the vehicle parameter matching degree verification is a verification process used to determine whether the traffic requirements corresponding to each trajectory segment are within the actual traffic capacity range of the autonomous driving light truck. Vehicle parameter matching degree verification can prevent the generation of traction trajectories that are mismatched with the light truck's body size, wheelbase, minimum turning radius, load status, or power performance.

[0130] Additionally, the vehicle parameter matching verification result is obtained by comparing the steering and traffic requirements corresponding to each trajectory segment with the overall vehicle performance parameters. The vehicle parameter matching verification result indicates whether the autonomous driving light truck can complete the driving actions corresponding to each trajectory segment based on its own vehicle performance.

[0131] Understandably, the corresponding steering requirements, passage space requirements, and driving requirements are determined based on the trajectory shape and road space of each trajectory segment. The vehicle dimensions, wheelbase, minimum turning radius, load status, and power performance of the autonomous driving light truck are determined from the overall vehicle performance parameters. The steering requirements corresponding to each trajectory segment are compared with the minimum turning radius. The vehicle dimensions and wheelbase are used to determine whether the road space can meet the vehicle's passage requirements. The load status and power performance are used to determine whether the vehicle has the ability to complete the corresponding driving maneuvers. The matching results of each trajectory segment are summarized to obtain the vehicle parameter matching verification results.

[0132] Step S315: Generate the pre-verification result based on the passage feasibility verification result, the collision risk verification result, and the vehicle parameter matching verification result; Understandably, the feasibility verification result, the collision risk verification result, and the vehicle parameter matching verification result are read respectively to determine whether the initial escape trajectory meets the road range and road connection requirements, whether there is no obstacle collision risk that would hinder the vehicle's passage, and whether it matches the overall performance of the autonomous driving light truck; when all verification results indicate that the initial escape trajectory meets the corresponding verification requirements, a pre-verification result indicating that the pre-verification has passed is generated; when any verification result indicates that the initial escape trajectory does not meet the corresponding verification requirements, a pre-verification result indicating that the pre-verification has failed is generated.

[0133] In one feasible implementation, when the pre-verification result is "pre-verification passed," the initial escape trajectory, the traffic feasibility verification result, the collision risk verification result, and the vehicle parameter matching verification result are sent to a manual verification terminal. The manual verification terminal displays the location of the initial escape trajectory on the target area map, the road traffic conditions corresponding to each trajectory segment, and the verification results, and receives the manual verification results input by the verification personnel for the initial escape trajectory.

[0134] It is understood that when the manual review result is "passed," it indicates that the initial escape trajectory, after algorithm pre-verification and manual review, can enter the vehicle-side safety assessment process, and the initial escape trajectory is determined as the verification escape trajectory. When the manual review result is "failed," the initial escape trajectory is regenerated according to the reason for failure recorded in the manual review result, or the corresponding trajectory segment in the initial escape trajectory is adjusted.

[0135] Step S316: When the pre-verification result is that the pre-verification is passed, determine the verification escape trajectory based on the vehicle parameter matching verification result.

[0136] It is understood that when the pre-verification result is that the pre-verification is passed, it indicates that the initial escape trajectory is feasible for road passage, there is no risk of collision with obstacles identified in the pre-verification stage, and the passage requirements corresponding to the initial escape trajectory are matched with the overall vehicle performance of the autonomous driving light truck; further, based on the vehicle parameter matching verification result, multiple trajectory segments that have passed the vehicle parameter matching degree verification and the arrangement relationship between the multiple trajectory segments are determined, and the trajectory segments that have passed the vehicle parameter matching degree verification and can be continuously connected according to the trajectory execution order are determined as the verified escape trajectory.

[0137] Step S32: Based on the vehicle's current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters, a safety assessment is performed on the verified escape trajectory to obtain the trajectory assessment result.

[0138] It should be noted that the current environmental perception data is real-time data collected by the vehicle-mounted LiDAR, millimeter-wave radar, and vision cameras regarding the surrounding environment of the autonomous light truck. This data is used to represent the actual situation of obstacles, road boundaries, and passable space around the vehicle, in order to determine whether the verified escape trajectory matches the actual environment.

[0139] Additionally, vehicle operating status data indicates the current power, braking, and steering status of the autonomous light truck. This data reflects whether the autonomous light truck currently possesses the conditions to perform driving, speed adjustment, steering, and braking actions according to the verified escape trajectory.

[0140] Furthermore, vehicle traffic performance parameters are data used to represent the vehicle performance of the autonomous light truck when navigating roads and completing turns. These parameters include at least the minimum turning radius of the autonomous light truck, used to determine whether the turning space and trajectory changes of the verified escape trajectory exceed the vehicle's actual traffic capacity.

[0141] Furthermore, the trajectory analysis result is obtained by comprehensively judging whether the verified escape trajectory can be safely executed, taking into account the on-site environment, the vehicle's current operating status, and the vehicle's traffic capacity. The trajectory analysis result is used to indicate whether the verified escape trajectory matches the current operating conditions and on-site environment of the autonomous driving light truck.

[0142] Understandably, based on the current environmental perception data, the distribution of obstacles, road boundaries, and passable space around the verification escape trajectory are identified, and it is determined whether the verification escape trajectory passes through an area occupied by obstacles or exceeds the passable area on site, thereby determining the environmental risk corresponding to the verification escape trajectory; based on the vehicle operating status data, it is determined whether the current power, braking, and steering status of the autonomous driving light truck can meet the driving requirements corresponding to the verification escape trajectory, thereby determining the matching status between the current vehicle operating condition and the verification escape trajectory; based on the vehicle's traffic performance parameters, it is determined whether the steering requirements and passable space of the verification escape trajectory meet the traffic capacity of the autonomous driving light truck, thereby determining the trajectory adaptation status of the verification escape trajectory; combining the environmental risk, the matching status between the current vehicle operating condition and the verification escape trajectory, and the trajectory adaptation status, the trajectory judgment result is obtained.

[0143] In one feasible implementation, step S32 may include steps S321 to S325: Step S321: Perform fusion processing on the current environmental perception data to obtain obstacle distribution data and passable area data; It should be noted that the obstacle distribution data is generated by comprehensively processing data collected by vehicle-mounted LiDAR, millimeter-wave radar, and vision cameras, and is used to represent the location and distribution of obstacles around the vehicle. The obstacle distribution data is used to determine whether the area traversed by the verification escape trajectory is occupied by obstacles, and whether there is a risk of collision with obstacles when the vehicle travels along the verification escape trajectory.

[0144] Additionally, the passable area data is determined based on the road boundaries around the vehicle, the distribution of obstacles, and the available road space, defining the areas where the autonomous light truck can travel. This passable area data is used to define the safe passage range for the vehicle and to determine whether the verification escape trajectory is within the passable range corresponding to the current environment.

[0145] It is understood that the environmental information collected by different vehicle-mounted sensing devices in the current environmental perception data is matched and integrated to identify the location of obstacles around the autonomous driving light truck and the road boundary; the obstacle distribution data is formed according to the location of the obstacles, and the area occupied by obstacles or unable to be passed by vehicles is excluded within the area defined by the road boundary, and the remaining area that can be driven by the autonomous driving light truck is determined as the passable area data.

[0146] Step S322: Determine the operating status of the vehicle power system, the operating status of the vehicle braking system, and the operating status of the vehicle steering system based on the vehicle operating status data; It should be noted that the vehicle powertrain operating status is status information used to indicate the current power output of the autonomous driving light truck. The vehicle powertrain operating status is used to determine whether the vehicle can currently provide the corresponding driving power according to the driving direction and driving requirements corresponding to the verified escape trajectory.

[0147] Additionally, the vehicle braking system operating status indicates whether the autonomous driving light truck's deceleration and stopping functions are functioning correctly. The vehicle braking system operating status is used to determine whether the vehicle can complete speed adjustment, deceleration, or stopping actions according to the trajectory requirements when executing the verification and extrication trajectory.

[0148] Furthermore, the vehicle steering system operating status is used to indicate the current steering function and steering execution status of the autonomous driving light truck. The vehicle steering system operating status is used to determine whether the vehicle can complete the corresponding steering actions according to the extension direction of the verified escape trajectory and the turning requirements.

[0149] It is understood that state information related to power output, braking execution, and steering execution is extracted from the vehicle operating status data. The operating status of the vehicle power system is determined based on the current operating status corresponding to the power output, the operating status of the vehicle braking system is determined based on the current operating status corresponding to the braking device, and the operating status of the vehicle steering system is determined based on the current operating status corresponding to the steering device. This provides a basis for subsequent judgment on whether the current operating condition of the vehicle matches the verified escape trajectory.

[0150] Step S323: Determine the trajectory curvature, turning space, and passage distance corresponding to each trajectory segment based on the vehicle traffic performance parameters and the verification escape trajectory; It should be noted that trajectory curvature is a trajectory feature used to represent the degree of curvature of a trajectory segment. Trajectory curvature can reflect the degree of steering change required for an autonomous light truck to pass through a corresponding trajectory segment, and is used to determine whether the corresponding trajectory segment meets the vehicle's minimum turning radius and other traffic performance requirements.

[0151] Additionally, turning space refers to the road space required for an autonomous light truck to complete a turn according to a corresponding trajectory segment. Turning space is related to the curvature of the trajectory segment, vehicle size, and vehicle turning capability, and is used to determine whether the passable area on site can meet the turning requirements of the autonomous light truck.

[0152] Furthermore, the clearance distance is the available distance between an autonomous light truck and the road boundary or surrounding obstacles when the truck travels along a corresponding trajectory segment. The clearance distance is used to determine whether an autonomous light truck has sufficient space to pass through a given trajectory segment.

[0153] Understandably, by analyzing the extension direction and curvature of each trajectory segment according to the arrangement order of the trajectory segments in the verification escape trajectory, the trajectory curvature corresponding to each trajectory segment is determined; by combining the vehicle traffic performance parameters and the steering changes corresponding to each trajectory segment, the turning space required for the autonomous driving light truck to pass through each trajectory segment is determined; and by further combining the vehicle traffic capacity and the road space around each trajectory segment, the passage distance corresponding to the autonomous driving light truck when traveling along each trajectory segment is determined.

[0154] Step S324: Based on the obstacle distribution data, the passable area data, the vehicle power system operating status, the vehicle braking system operating status, the vehicle steering system operating status, the trajectory curvature, the turning space, and the passage distance, determine the environmental risk data, vehicle operating condition matching data, and trajectory adaptation data. It should be noted that environmental risk data is used to indicate whether there is a safety conflict between the verified escape trajectory and the current surrounding environment of the autonomous driving light truck. Environmental risk data mainly reflects whether the trajectory segment passes through areas occupied by obstacles, whether it exceeds the passable area, and whether there is insufficient passage space.

[0155] Additionally, vehicle condition matching data indicates whether the current power, braking, and steering operation of the autonomous driving light truck can meet the execution requirements of the verification escape trajectory. Vehicle condition matching data is used to determine whether the vehicle's current operating state supports completing driving, steering, speed adjustment, and braking according to the verification escape trajectory.

[0156] Furthermore, trajectory adaptation data is used to indicate whether the trajectory characteristics and spatial requirements of the verified escape trajectory are compatible with the traffic capacity of the autonomous light truck. The trajectory adaptation data mainly reflects whether the curvature, turning space, and spacing of each trajectory segment meet the vehicle's traffic performance requirements.

[0157] Understandably, each trajectory segment is mapped to the obstacle distribution data and the passable area data to determine whether each trajectory segment passes through the obstacle area and whether it is within the passable area, thereby determining the environmental risk data; the driving, braking, and steering requirements corresponding to each trajectory segment are compared with the vehicle power system operating status, the vehicle braking system operating status, and the vehicle steering system operating status, respectively, thereby determining the vehicle operating condition matching data; the trajectory curvature, the turning space, and the passage distance are compared with the vehicle passage performance parameters reflecting the vehicle's passage capacity, thereby determining the trajectory adaptation data.

[0158] Step S325: Based on the environmental risk data, the vehicle operating condition matching data, and the trajectory adaptation data, a comprehensive safety assessment is performed to obtain the trajectory assessment result.

[0159] It should be noted that the trajectory analysis result is formed by comprehensively judging the verified escape trajectory from three aspects: on-site environmental risks, the current operating condition of the vehicle, and the compatibility of the trajectory with the vehicle's traffic capacity. The trajectory analysis result is used to indicate whether the verified escape trajectory meets the safety execution requirements corresponding to the current vehicle state and on-site environment.

[0160] It is understandable that the environmental risk data is determined to indicate that the verification escape trajectory has an obstacle collision risk or exceeds the passable area; the vehicle operating condition matching data is determined to indicate that the vehicle's current power, braking and steering operation can meet the trajectory execution requirements; and the trajectory adaptation data is determined to indicate that the trajectory curvature, turning space and passage distance of each trajectory segment meet the vehicle's passage performance requirements. By combining the results of each judgment, it is determined whether the verification escape trajectory meets the preset safety conditions, and the trajectory analysis result is obtained.

[0161] This embodiment provides a method for generating an escape trajectory for an autonomous vehicle. First, the initial escape trajectory is pre-verified to determine its feasibility, obstacle collision risk, and vehicle parameter matching. A verification escape trajectory is then determined. Next, the vehicle's current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters are combined to conduct a safety assessment of the verification escape trajectory. This verifies the escape trajectory from both pre-planned and real-time on-site conditions, ensuring that the trajectory assessment results reflect the actual execution safety and operational adaptability of the verification escape trajectory.

[0162] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the method for generating the escape trajectory of autonomous vehicles in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0163] This application also provides a device for generating an escape trajectory for an autonomous vehicle, please refer to... Figure 3 The autonomous vehicle's escape trajectory generation device includes: The obstacle avoidance detection module 10 is used to acquire vehicle obstacle data, target area map data, road topology data, vehicle performance parameters and global road condition data when it detects that the autonomous driving light truck is stuck. The trajectory generation module 20 is used to generate an initial escape trajectory based on the vehicle entrapment data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data. The trajectory verification module 30 is used to verify and assess the safety of the initial escape trajectory, and obtain the trajectory assessment result. The trajectory generation module 20 is also used to generate a target escape trajectory based on the trajectory analysis results.

[0164] The autonomous vehicle traction trajectory generation device provided in this application, employing the autonomous vehicle traction trajectory generation method in the above embodiments, can solve the technical problem that existing L4-level autonomous light trucks have low efficiency in determining traction control strategies when stuck on structured roads, and are difficult to match with the vehicle's current state and on-site environmental conditions. Compared with the prior art, the beneficial effects of the autonomous vehicle traction trajectory generation device provided in this application are the same as those of the autonomous vehicle traction trajectory generation method provided in the above embodiments, and other technical features in the autonomous vehicle traction trajectory generation device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0165] In one embodiment, the trajectory verification module 30 is further configured to pre-verify the initial escape trajectory, obtain a pre-verification result, and determine the verification escape trajectory based on the pre-verification result; Based on the vehicle's current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters, a safety assessment is performed on the verified escape trajectory to obtain the trajectory assessment result.

[0166] In one embodiment, the trajectory verification module 30 is further configured to parse the initial escape trajectory to obtain multiple trajectory segments arranged in the order of trajectory execution; Based on the trajectory segment, the target area map data, and the road topology data, the feasibility of the trajectory is verified to obtain the feasibility verification result. Based on the trajectory segment and the global road condition data, an obstacle collision risk verification is performed to obtain the collision risk verification result. Based on the trajectory segment and the vehicle performance parameters, the vehicle parameter matching degree is verified to obtain the vehicle parameter matching verification result. The pre-verification result is generated based on the passage feasibility verification result, the collision risk verification result, and the vehicle parameter matching verification result. When the pre-verification result is "pre-verification passed", the verification escape trajectory is determined based on the vehicle parameter matching verification result.

[0167] In one embodiment, the trajectory verification module 30 is further configured to perform fusion processing on the current environmental perception data to obtain obstacle distribution data and passable area data; The operating status of the vehicle powertrain system, the vehicle braking system, and the vehicle steering system are determined based on the vehicle operating status data. Based on the vehicle traffic performance parameters and the verification escape trajectory, determine the trajectory curvature, turning space and passage distance corresponding to each trajectory segment; Based on the obstacle distribution data, the passable area data, the vehicle power system operating status, the vehicle braking system operating status, the vehicle steering system operating status, the trajectory curvature, the turning space, and the passage distance, environmental risk data, vehicle operating condition matching data, and trajectory adaptation data are determined. A comprehensive safety assessment is conducted based on the environmental risk data, the vehicle operating condition matching data, and the trajectory adaptation data to obtain the trajectory assessment result.

[0168] In one embodiment, the trajectory generation module 20 is further configured to parse the vehicle entrapment data to obtain the current position, current vehicle posture, current chassis condition, and current power status of the autonomous driving light truck. Candidate traffic areas and candidate trajectory nodes within the candidate traffic areas are determined based on the current location, the target area map data, the road topology data, and the global traffic data. Based on the vehicle performance parameters, the current vehicle posture, the current chassis condition, and the current power state, determine the trajectory curvature constraint, turning radius constraint, driving speed constraint, and braking distance constraint; A constrained escape trajectory is generated based on the candidate trajectory nodes, the trajectory curvature constraint, the turning radius constraint, the driving speed constraint, and the braking distance constraint; The initial escape trajectory is determined based on the trajectory length, trajectory continuity, road occupancy, and obstacle distribution of each constrained escape trajectory.

[0169] In one embodiment, the trajectory generation module 20 is further configured to parse the trajectory assessment results to obtain the trajectory safety status, trajectory risk type, and risk trajectory segment; When the trajectory safety status is that the verified escape trajectory meets the preset safety conditions, the verified escape trajectory is determined as the target escape trajectory; When the trajectory safety status is that the verified escape trajectory does not meet the preset safety conditions, trajectory replanning constraints are generated based on the trajectory risk type, the risk trajectory segment, the vehicle entrapment data, and the global road condition data. Based on the trajectory replanning constraints, the trajectory points, driving direction, and driving speed corresponding to the risk trajectory segment are adjusted to obtain the replanned escape trajectory; The replanned escape trajectory is verified and its safety is assessed to obtain an updated trajectory assessment result. When the updated trajectory assessment result indicates that the replanned escape trajectory meets the preset safety conditions, the replanned escape trajectory is determined as the target escape trajectory.

[0170] In one embodiment, the trajectory generation module 20 is further configured to determine the execution trajectory segment and the target vehicle state corresponding to the execution trajectory segment based on the target escape trajectory; Based on the execution trajectory segment and the target vehicle state, target driving parameters, target steering parameters, and target braking parameters are generated. The autonomous driving light truck is controlled to perform an escape operation based on the target driving parameters, target steering parameters, and target braking parameters. The vehicle position, body posture, driving speed, and surrounding environment change data of the autonomous driving light truck are also acquired. The escape execution status is determined based on the vehicle position, vehicle posture, driving speed, and changes in the surrounding environment. When the autonomous driving light truck deviates from its trajectory during the escape execution state, the unexecuted trajectory segment is corrected, and an updated target escape trajectory is generated.

[0171] This application provides an autonomous vehicle traction trajectory generation device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the autonomous vehicle traction trajectory generation method in Embodiment 1 above.

[0172] The following is for reference. Figure 4The diagram illustrates a structural schematic of an autonomous vehicle traction trajectory generation device suitable for implementing embodiments of this application. The autonomous vehicle traction trajectory generation device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), tablets, PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The autonomous vehicle escape trajectory generation device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0173] like Figure 4 As shown, the autonomous vehicle's obstacle avoidance trajectory generation device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the autonomous vehicle's obstacle avoidance trajectory generation device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the autonomous vehicle's obstacle avoidance trajectory generation device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows an autonomous vehicle's obstacle avoidance trajectory generation device with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems may be implemented alternatively.

[0174] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0175] The autonomous vehicle traction trajectory generation device provided in this application, employing the autonomous vehicle traction trajectory generation method described in the above embodiments, can solve the technical problem that existing L4-level autonomous light trucks have low efficiency in determining traction control strategies when stuck on structured roads, and are difficult to match with the vehicle's current state and on-site environmental conditions. Compared with the prior art, the beneficial effects of the autonomous vehicle traction trajectory generation device provided in this application are the same as those of the autonomous vehicle traction trajectory generation method provided in the above embodiments, and other technical features in this autonomous vehicle traction trajectory generation device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0176] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0177] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0178] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the autonomous vehicle obstacle avoidance trajectory generation method in the above embodiments.

[0179] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), Erasable Programmable Read Only Memory (EPROM), optical fiber, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0180] The aforementioned computer-readable storage medium may be included in the autonomous vehicle's trajectory generation device for escaping difficulties; or it may exist independently and not be installed in the autonomous vehicle's trajectory generation device for escaping difficulties.

[0181] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by the autonomous vehicle's traction trajectory generation device, the device causes the autonomous vehicle's traction trajectory generation device to: upon detecting that an autonomous light truck has become stuck, acquire vehicle traction data, target area map data, road topology data, vehicle performance parameters, and global road condition data; generate an initial traction trajectory based on the vehicle traction data, target area map data, road topology data, vehicle performance parameters, and global road condition data; verify and assess the initial traction trajectory to obtain a trajectory assessment result; and generate a target traction trajectory based on the trajectory assessment result.

[0182] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0183] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0184] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0185] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for generating an escape trajectory for an autonomous vehicle. This solves the technical problem that existing L4-level autonomous light trucks, when stuck on structured roads, have low efficiency in determining their escape control strategies and difficulty in matching them with the vehicle's current state and the surrounding environment. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the autonomous vehicle escape trajectory generation method provided in the above embodiments, and will not be repeated here.

[0186] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for generating an escape trajectory for an autonomous vehicle.

[0187] The computer program product provided in this application can solve the technical problem that existing L4-level autonomous driving light trucks have low efficiency in determining the escape control strategy when stuck on structured roads, and it is difficult to match the current state of the vehicle and the on-site environmental conditions. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the autonomous vehicle escape trajectory generation method provided in the above embodiments, and will not be repeated here.

[0188] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for generating an escape trajectory for an autonomous vehicle, characterized in that, The method includes: When an autonomous light truck is detected to be stuck, the system acquires vehicle stuck data, target area map data, road topology data, vehicle performance parameters, and global road condition data. An initial escape trajectory is generated based on the vehicle entanglement data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data; The initial escape trajectory is verified and safety is assessed to obtain the trajectory assessment result; Based on the trajectory analysis results, a target escape trajectory is generated.

2. The method as described in claim 1, characterized in that, The steps of verifying and assessing the initial escape trajectory to obtain the trajectory assessment result include: The initial escape trajectory is pre-verified to obtain the pre-verification result, and the verified escape trajectory is determined based on the pre-verification result; Based on the vehicle's current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters, a safety assessment is performed on the verified escape trajectory to obtain the trajectory assessment result.

3. The method as described in claim 2, characterized in that, The steps of pre-verifying the initial escape trajectory, obtaining the pre-verification result, and determining the verified escape trajectory based on the pre-verification result include: The initial escape trajectory is analyzed to obtain multiple trajectory segments arranged in the order of trajectory execution; Based on the trajectory segment, the target area map data, and the road topology data, the feasibility of the trajectory is verified to obtain the feasibility verification result. Based on the trajectory segment and the global road condition data, an obstacle collision risk verification is performed to obtain the collision risk verification result. Based on the trajectory segment and the vehicle performance parameters, the vehicle parameter matching degree is verified to obtain the vehicle parameter matching verification result. The pre-verification result is generated based on the passage feasibility verification result, the collision risk verification result, and the vehicle parameter matching verification result. When the pre-verification result is "pre-verification passed", the verification escape trajectory is determined based on the vehicle parameter matching verification result.

4. The method as described in claim 2, characterized in that, The step of performing a safety assessment of the verified escape trajectory based on the vehicle's current environmental perception data, vehicle operating status data, and vehicle traffic performance parameters, and obtaining the trajectory assessment result, includes: The current environmental perception data is fused to obtain obstacle distribution data and passable area data; The operating status of the vehicle powertrain system, the vehicle braking system, and the vehicle steering system are determined based on the vehicle operating status data. Based on the vehicle traffic performance parameters and the verification escape trajectory, determine the trajectory curvature, turning space and passage distance corresponding to each trajectory segment; Based on the obstacle distribution data, the passable area data, the vehicle power system operating status, the vehicle braking system operating status, the vehicle steering system operating status, the trajectory curvature, the turning space, and the passage distance, environmental risk data, vehicle operating condition matching data, and trajectory adaptation data are determined. A comprehensive safety assessment is conducted based on the environmental risk data, the vehicle operating condition matching data, and the trajectory adaptation data to obtain the trajectory assessment result.

5. The method as described in claim 1, characterized in that, The step of generating an initial escape trajectory based on the vehicle entanglement data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data includes: The vehicle entrapment data is analyzed to obtain the current position, current vehicle posture, current chassis condition, and current power status of the autonomous driving light truck. Candidate traffic areas and candidate trajectory nodes within the candidate traffic areas are determined based on the current location, the target area map data, the road topology data, and the global traffic data. Based on the vehicle performance parameters, the current vehicle posture, the current chassis condition, and the current power state, determine the trajectory curvature constraint, turning radius constraint, driving speed constraint, and braking distance constraint; A constrained escape trajectory is generated based on the candidate trajectory nodes, the trajectory curvature constraint, the turning radius constraint, the driving speed constraint, and the braking distance constraint; The initial escape trajectory is determined based on the trajectory length, trajectory continuity, road occupancy, and obstacle distribution of each constrained escape trajectory.

6. The method as described in claim 1, characterized in that, The step of generating the target escape trajectory based on the trajectory analysis results includes: By analyzing the trajectory analysis results, the trajectory safety status, trajectory risk type, and risk trajectory segment can be obtained; When the trajectory safety status is that the verified escape trajectory meets the preset safety conditions, the verified escape trajectory is determined as the target escape trajectory; When the trajectory safety status is that the verified escape trajectory does not meet the preset safety conditions, trajectory replanning constraints are generated based on the trajectory risk type, the risk trajectory segment, the vehicle entrapment data, and the global road condition data. Based on the trajectory replanning constraints, the trajectory points, driving direction, and driving speed corresponding to the risk trajectory segment are adjusted to obtain the replanned escape trajectory; The replanned escape trajectory is verified and its safety is assessed to obtain an updated trajectory assessment result. When the updated trajectory assessment result indicates that the replanned escape trajectory meets the preset safety conditions, the replanned escape trajectory is determined as the target escape trajectory.

7. The method as described in claim 1, characterized in that, After the step of generating the target escape trajectory based on the trajectory analysis results, the method further includes: The execution trajectory segment and the corresponding target vehicle status are determined based on the target escape trajectory. Based on the execution trajectory segment and the target vehicle state, target driving parameters, target steering parameters, and target braking parameters are generated. The autonomous driving light truck is controlled to perform an escape operation based on the target driving parameters, target steering parameters, and target braking parameters. The vehicle position, body posture, driving speed, and surrounding environment change data of the autonomous driving light truck are also acquired. The escape execution status is determined based on the vehicle position, vehicle posture, driving speed, and changes in the surrounding environment. When the autonomous driving light truck deviates from its trajectory during the escape execution state, the unexecuted trajectory segment is corrected, and an updated target escape trajectory is generated.

8. A trajectory device for escaping difficulties in an autonomous vehicle, characterized in that, The device includes: The obstacle avoidance detection module is used to acquire vehicle obstacle data, target area map data, road topology data, vehicle performance parameters, and global road condition data when an autonomous driving light truck is detected to be stuck. The trajectory generation module is used to generate an initial escape trajectory based on the vehicle entrapment data, the target area map data, the road topology data, the vehicle performance parameters, and the global road condition data. The trajectory verification module is used to verify and assess the safety of the initial escape trajectory, and obtain the trajectory assessment result. The trajectory generation module is also used to generate a target escape trajectory based on the trajectory analysis results.

9. A trajectory tracking device for escaping difficulties in an autonomous vehicle, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the autonomous vehicle trajectories method as claimed in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the autonomous vehicle's escape trajectory method as described in any one of claims 1 to 7.

Images