Method and device for determining driving track
By smoothing the initial planned trajectory and implementing a safety fallback plan, combined with local map information and mathematical optimization algorithms, the optimal executable trajectory is generated. This solves the problems of unsmooth trajectories and rule violations in data-driven models during autonomous driving, thereby improving the safety and traffic efficiency of autonomous driving.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA FAW CO LTD
- Filing Date
- 2026-03-30
- Publication Date
- 2026-05-08
AI Technical Summary
Existing data-driven decision planning models based on machine learning/deep learning have problems in autonomous driving, such as high dependence on training data, uneven output trajectory, failure to meet vehicle dynamics constraints, and potential violation of traffic rules in rare scenarios, leading to a decline in safety, comfort, and traffic efficiency.
By smoothing the initial planned trajectory, a safe fallback planned trajectory is generated, and arbitration evaluation is performed. Combining local map information and mathematical optimization algorithms, the optimal executable trajectory is generated.
It improves the safety, comfort, and accessibility of autonomous driving, ensures that trajectories conform to vehicle dynamics constraints and traffic rules, and enhances decision-making capabilities in rare scenarios.
Smart Images
Figure CN121989999A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of autonomous driving technology, and in particular to a method and apparatus for determining a driving trajectory. Background Technology
[0002] With the development of artificial intelligence technology, the field of autonomous driving has begun to widely adopt data-driven methods based on machine learning / deep learning, especially end-to-end decision-making and planning models. Typically, these models are trained with massive amounts of road test data and can directly output planned trajectories based on perceived map information. They have advantages such as lightweight code, large planning space, and good imitation of human driving behavior.
[0003] However, these data-driven decision-making and planning models also face some inherent limitations: First, their performance is highly dependent on the quantity, quality, and scenario coverage of training data, making it difficult to effectively incorporate prior knowledge such as vehicle dynamics and traffic rules. Second, in real-world applications, the model may output trajectories with abrupt curvature changes, lack of smoothness, or failure to conform to vehicle dynamics constraints. Furthermore, when faced with rare or extreme scenarios with insufficient training data coverage, these models may produce unreasonable plans that violate traffic rules or common safety sense. These unreasonable plans leading to adverse behaviors severely impact the safety, comfort, and traffic efficiency of autonomous vehicles.
[0004] Currently, post-processing and security enhancement techniques for the output of decision-programming models are not yet mature. Simple trajectory smoothing methods may not be able to handle fundamental errors in the model output, while relying entirely on traditional rule-based methods negates the advantages of decision-programming models. Summary of the Invention
[0005] In view of the above, embodiments of this application provide at least one method and apparatus for determining a driving trajectory to overcome at least one of the above-mentioned defects.
[0006] In a first aspect, an exemplary embodiment of this application provides a method for determining a driving trajectory, the method comprising: acquiring at least one initial planned trajectory output by a decision planning model for a target vehicle; The initial planned trajectory is smoothed to obtain at least one model output trajectory; Based on the acquired local map related to the target vehicle, at least one safety fallback planning trajectory for the target vehicle is generated using the homotopy trajectory generation method. The model output trajectory and the safety fallback planned trajectory are arbitrated and evaluated, and the optimal trajectory after evaluation is taken as the executable trajectory of the current target vehicle.
[0007] Optionally, the step of smoothing the initial planned trajectory to obtain at least one model output trajectory includes: For each initial planned trajectory, the target sampling points corresponding to the initial planned trajectory are obtained after filtering the initial planned trajectory. Using a pre-built penalty function, the target sampling points corresponding to the initial planned trajectory are optimized to obtain the model output trajectory corresponding to the model output trajectory.
[0008] Optionally, the step of generating at least one safety fallback planning trajectory based on the acquired local map related to the target vehicle using a homotopy trajectory generation method includes: Obtain a local map containing static and dynamic obstacle information related to the target vehicle; Based on the drivable area segmented by obstacles in the local map, lateral and longitudinal decisions are made. By combining the lateral and vertical decisions, at least one homotopic traversal decision is generated; Based on the homotopy traffic decision, a mathematical optimization algorithm is invoked to generate at least one safe fallback planning trajectory for the target vehicle.
[0009] Optionally, the step of arbitrarily evaluating the model output trajectory and the safety fallback planning trajectory, and using the evaluated optimal trajectory as the executable trajectory of the current target vehicle, includes: The model output trajectory and the safety fallback planning trajectory are respectively input into the target evaluator to obtain the evaluation score of each model output trajectory and each safety fallback planning trajectory; Based on the evaluation score, determine whether the trajectory meets the execution requirements; If the conditions are met, the trajectory with the lowest evaluation score among the qualified trajectories will be determined as the executable trajectory for the current target vehicle.
[0010] Optionally, the determination method further includes: if the condition is not met, then the executable trajectory determined at the previous sampling time is combined with the current driving data of the target vehicle for fitting, and the fitted trajectory is determined as the current executable trajectory of the target vehicle.
[0011] Secondly, embodiments of this application also provide a device for determining a driving trajectory, the device comprising: The acquisition module acquires at least one initial planned trajectory output by the decision planning model for the target vehicle. The smoothing module smooths the initial planned trajectory to obtain at least one model output trajectory. The generation module generates at least one safety fallback planning trajectory for the target vehicle based on the acquired local map related to the target vehicle using the homotopy trajectory generation method. The evaluation module arbitrates and evaluates the model output trajectory and the safety fallback planning trajectory, and uses the optimal trajectory after evaluation as the executable trajectory of the current target vehicle.
[0012] Optionally, the smoothing module is specifically used for: For each initial planned trajectory, the target sampling points corresponding to the initial planned trajectory are obtained after filtering the initial planned trajectory. Using a pre-built penalty function, the target sampling points corresponding to the initial planned trajectory are optimized to obtain the model output trajectory corresponding to the model output trajectory.
[0013] Optionally, the generating module is specifically used for: Obtain a local map containing static and dynamic obstacle information related to the target vehicle; Based on the drivable area segmented by obstacles in the local map, lateral and longitudinal decisions are made. By combining the lateral and vertical decisions, at least one homotopic traversal decision is generated; Based on the homotopy traffic decision, a mathematical optimization algorithm is invoked to generate at least one safe fallback planning trajectory for the target vehicle.
[0014] Optionally, the evaluation module is specifically used for: The model output trajectory and the safety fallback planning trajectory are respectively input into the target evaluator to obtain the evaluation score of each model output trajectory and each safety fallback planning trajectory; Based on the evaluation score, determine whether the trajectory meets the execution requirements; If the conditions are met, the trajectory with the lowest evaluation score among the qualified trajectories will be determined as the executable trajectory for the current target vehicle.
[0015] Optionally, the evaluation module is further configured to: if the conditions are not met, fit the executable trajectory determined at the previous sampling time with the current driving data of the target vehicle, and determine the fitted trajectory as the current executable trajectory of the target vehicle.
[0016] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the above-described method for determining the driving trajectory.
[0017] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the above-described method for determining the driving trajectory.
[0018] A method and apparatus for determining a driving trajectory, as proposed in the exemplary embodiments of this application, can greatly improve the safety, comfort, and accessibility of autonomous driving.
[0019] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description
[0020] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 A flowchart illustrating a method for determining a driving trajectory provided by an exemplary embodiment of this application; Figure 2 A schematic diagram of the structure of the driving trajectory determination device provided in an exemplary embodiment of this application is shown; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an exemplary embodiment of this application. Detailed Implementation
[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.
[0023] The terms “a,” “an,” “the,” and “the” are used in this specification to indicate the presence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended inclusion and to mean that there may be other elements / components / etc. in addition to the listed elements / components / etc.; the terms “first” and “second” are used only as markings and are not a limitation on the number of objects.
[0024] It should be understood that in the embodiments of this application, "at least one" means one or more, and "more than one" means two or more. "And / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The character " / " generally indicates that the related objects before and after it are in an "or" relationship. "Contains A, B and / or C" means containing any one, two, or three of A, B, and C.
[0025] It should be understood that in the embodiments of this application, "B corresponding to A", "B corresponding to A", "A corresponds to B" or "B corresponds to A" means that B is associated with A, and B can be determined based on A. Determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.
[0026] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0027] With the development of artificial intelligence technology, the field of autonomous driving has begun to widely adopt data-driven methods based on machine learning / deep learning, especially end-to-end decision-making and planning models. Typically, these models are trained with massive amounts of road test data and can directly output planned trajectories based on perceived map information. They have advantages such as lightweight code, large planning space, and good imitation of human driving behavior.
[0028] However, these data-driven decision-making and planning models also face some inherent limitations: First, their performance is highly dependent on the quantity, quality, and scenario coverage of training data, making it difficult to effectively incorporate prior knowledge such as vehicle dynamics and traffic rules. Second, in real-world applications, the model may output trajectories with abrupt curvature changes, lack of smoothness, or failure to conform to vehicle dynamics constraints. Furthermore, when faced with rare or extreme scenarios with insufficient training data coverage, these models may produce unreasonable plans that violate traffic rules or common safety sense. These unreasonable plans leading to adverse behaviors severely impact the safety, comfort, and traffic efficiency of autonomous vehicles.
[0029] Currently, post-processing and security enhancement techniques for the output of decision-programming models are not yet mature. Simple trajectory smoothing methods may not be able to handle fundamental errors in the model output, while relying entirely on traditional rule-based methods negates the advantages of decision-programming models.
[0030] For the reasons mentioned above, it is clear that existing methods for determining driving trajectories cannot meet people's needs for accuracy.
[0031] To address at least one of the problems mentioned above, this application proposes a method and apparatus for determining a driving trajectory. For ease of understanding, the method and apparatus for determining a driving trajectory according to embodiments of this application will be described in detail below.
[0032] Please see Figure 1 This diagram illustrates a flowchart of a method for determining a driving trajectory provided by an exemplary embodiment of this application. Figure 1 As shown, the method for determining the driving trajectory in an exemplary embodiment of this application specifically includes: Step S101: Obtain at least one initial planning trajectory output by the decision planning model for the target vehicle.
[0033] In an exemplary embodiment of this application, the decision planning model can be an intelligent model in the vehicle motion control system, used to generate at least one planned driving trajectory for the vehicle during autonomous driving. During autonomous driving, the vehicle motion control system can perform driving trajectory planning at predetermined intervals to obtain at least one initial planned trajectory.
[0034] Step S102: Smooth the initial planned trajectory to obtain at least one model output trajectory.
[0035] In the exemplary embodiment of this application, step S102 may include the following steps in actual implementation: Step S1021: For each initial planned trajectory, after filtering the initial planned trajectory, the target sampling points corresponding to the initial planned trajectory are obtained.
[0036] Specifically, after obtaining the initial planned trajectory, for each initial planned trajectory, the initial planned trajectory can be used as the initial sampling point for filtering. Here, the filtering process may include: first removing duplicate invalid points and / or abnormal points, and then reselecting points according to a preset time step, thereby obtaining the target sampling point corresponding to the initial planned trajectory.
[0037] Step S1022: Use the pre-built penalty function to optimize the target sampling points corresponding to the initial planned trajectory to obtain the model output trajectory corresponding to the model output trajectory.
[0038] Specifically, the penalty function can be constructed using a trajectory tracking penalty function, a vehicle control efficiency function, a comfort function, and a safety limit function. Furthermore, by fully utilizing each of these functions, the target sampling points corresponding to each initial planned trajectory are smoothed to ensure that the smoothed trajectory curvature achieves fourth-order continuity, closely matches the corresponding initial planned trajectory, and satisfies vehicle motion constraints, vehicle comfort, and safety limits.
[0039] Return to reference Figure 1 Step S103: Based on the acquired local map related to the target vehicle, generate at least one safety fallback planning trajectory for the target vehicle using the homotopy trajectory generation method.
[0040] Here, the homotopy trajectory generation method is a trajectory generation method based on geometric decision-making and mathematical optimization. The concept of homotopy refers to a "channel" with complete connectivity in planar space. For example, if there is an obstacle in the center of a road, then going around the obstacle to the left and going around the obstacle to the right each form a homotopy channel. In the exemplary embodiment of this application, regarding step S103, in actual implementation, it may include the following steps: Step S1031: Obtain a local map containing static and dynamic obstacle information related to the target vehicle.
[0041] In an exemplary embodiment of this application, when the target vehicle is driving autonomously, a local map containing static and dynamic obstacle information related to the current target vehicle can be periodically acquired. Here, the local map may be a portion of the map centered on the target vehicle and cut off according to a predetermined range. After the local map is determined, static and dynamic obstacles in the area covered by the local map can be projected onto the local map.
[0042] Step S1032: Based on the drivable area divided by obstacles in the local map, make lateral and longitudinal decisions.
[0043] After static and dynamic obstacles in the area covered by the local map are projected onto the local map, the static and dynamic obstacles can divide the local map into drivable areas. Then, lateral and longitudinal decisions can be made for the drivable areas divided by obstacles in the local map.
[0044] Here, lateral decision-making refers to the target vehicle's driving strategy in the horizontal direction of the local map, such as detouring in the left lane, detouring in the right lane, changing lanes, etc.; longitudinal strategy refers to the target vehicle's driving strategy in the vertical direction of the local map, such as overtaking, yielding, ignoring, etc.
[0045] Step S1033: Combine the lateral and longitudinal decisions to generate at least one homotopic traversal decision.
[0046] Specifically, after obtaining the lateral and longitudinal decisions, a pre-defined fusion judgment strategy based on the homotopy trajectory generation method is used to process them and generate at least one homotopy travel decision.
[0047] Step S1034: Based on the homotopy traffic decision, call a mathematical optimization algorithm to generate at least one safe fallback planning trajectory for the target vehicle.
[0048] As an example, the mathematical optimization algorithm could be the MPC model predictive control mathematical optimization method. After obtaining the general traffic decision, the MPC model predictive control mathematical optimization method can be called to calculate at least one safe fallback trajectory for the target vehicle from multiple homotopic traffic decisions.
[0049] Return to reference Figure 1 Step S104: Arbitrate and evaluate the model output trajectory and the safety fallback planning trajectory, and use the optimal trajectory after evaluation as the executable trajectory of the current target vehicle.
[0050] Specifically, step S104, in its implementation, may include the following steps: Step S1041: Input the model output trajectory and the safety fallback planning trajectory into the target evaluator to obtain the evaluation score of each model output trajectory and each safety fallback planning trajectory.
[0051] Here, the target evaluator can be a trajectory evaluator, which takes the model output trajectory and the safety fallback planning trajectory as input values and gives an evaluation score for each model output trajectory and each safety fallback planning trajectory.
[0052] Step S1042: Determine whether the trajectory meets the execution requirements based on the evaluation score.
[0053] Here, since the evaluation score must meet a safety threshold for the trajectory to be considered executable, a safety threshold can be preset. If all evaluation scores are below the safety threshold, the execution requirement is considered not met. If the evaluation score is not lower than the safety threshold, the execution requirement is considered met.
[0054] Step S1043: If the condition is met, the trajectory with the lowest evaluation score among the qualified trajectories is determined as the executable trajectory of the current target vehicle.
[0055] In addition, in step S1044, if the condition is not met, the executable trajectory determined at the previous sampling time is fitted together with the current driving data of the target vehicle. In step S1045, the fitted trajectory is determined as the executable trajectory of the current target vehicle.
[0056] Here, the current driving data of the target vehicle may include: the current speed and direction angle of the target vehicle. The relevant driving data can be fitted with the executable trajectory determined at the previous sampling time to obtain the current executable trajectory of the target vehicle.
[0057] The driving trajectory determination method proposed according to the exemplary embodiments of this application can greatly improve the safety, comfort and accessibility of autonomous driving.
[0058] Based on the same application concept, this application also provides an apparatus corresponding to the method provided in the above embodiments. Since the principle of the apparatus in this application to solve the problem is similar to the determination method in the above embodiments of this application, the implementation of the apparatus can refer to the implementation of the method, and the repeated parts will not be described again.
[0059] Figure 2 A schematic diagram of the structure of a driving trajectory determination device provided for an exemplary embodiment of this application.
[0060] Module 210 acquires at least one initial planned trajectory output by the decision planning model for the target vehicle; The smoothing module 220 smooths the initial planned trajectory to obtain at least one model output trajectory. The generation module 230 generates at least one safety fallback planning trajectory for the target vehicle based on the acquired local map related to the target vehicle using a homotopy trajectory generation method. The evaluation module 240 performs arbitration evaluation on the model output trajectory and the safety fallback planning trajectory, and uses the optimal trajectory after evaluation as the executable trajectory of the current target vehicle.
[0061] Optionally, the smoothing module 220 is specifically used for: For each initial planned trajectory, the target sampling points corresponding to the initial planned trajectory are obtained after filtering the initial planned trajectory. Using a pre-built penalty function, the target sampling points corresponding to the initial planned trajectory are optimized to obtain the model output trajectory corresponding to the model output trajectory.
[0062] Optionally, the generating module 230 is specifically used for: Obtain a local map containing static and dynamic obstacle information related to the target vehicle; Based on the drivable area segmented by obstacles in the local map, lateral and longitudinal decisions are made. By combining the lateral and vertical decisions, at least one homotopic traversal decision is generated; Based on the homotopy traffic decision, a mathematical optimization algorithm is invoked to generate at least one safe fallback planning trajectory for the target vehicle.
[0063] Optionally, the evaluation module 230 is specifically used for: The model output trajectory and the safety fallback planning trajectory are respectively input into the target evaluator to obtain the evaluation score of each model output trajectory and each safety fallback planning trajectory; Based on the evaluation score, determine whether the trajectory meets the execution requirements; If the conditions are met, the trajectory with the lowest evaluation score among the qualified trajectories will be determined as the executable trajectory for the current target vehicle.
[0064] Optionally, the evaluation module 230 is further configured to: if the condition is not met, combine the executable trajectory determined at the previous sampling time with the current driving data of the target vehicle to fit the current trajectory, and determine the fitted trajectory as the current executable trajectory of the target vehicle.
[0065] The determining device according to the embodiments of this application can greatly improve the safety, comfort and accessibility of autonomous driving.
[0066] Please see Figure 3 , Figure 3 A schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this application. For example... Figure 3 As shown, the electronic device 300 includes a processor 310, a memory 320, and a bus 330.
[0067] The memory 320 stores machine-readable instructions that can be executed by the processor 310. When the electronic device 300 is running, the processor 310 and the memory 320 communicate via the bus 330. When the machine-readable instructions are executed by the processor 310, the steps of the driving trajectory determination method in any of the above embodiments can be performed.
[0068] This application also provides a computer-readable storage medium storing a computer program that, when run by a processor, can execute the steps of the driving trajectory determination method as described in any of the above embodiments.
[0069] The computer-readable storage medium described in this application embodiment can greatly improve the safety, comfort, and accessibility of autonomous driving.
[0070] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.
[0071] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0072] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.
[0073] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0074] The above are merely specific embodiments 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.
Claims
1. A method for determining a driving trajectory, characterized in that, The determination method includes: Obtain at least one initial planned trajectory output by the decision planning model for the target vehicle; The initial planned trajectory is smoothed to obtain at least one model output trajectory; Based on the acquired local map related to the target vehicle, at least one safety fallback planning trajectory for the target vehicle is generated using the homotopy trajectory generation method. The model output trajectory and the safety fallback planned trajectory are arbitrated and evaluated, and the optimal trajectory after evaluation is taken as the executable trajectory of the current target vehicle.
2. The determination method according to claim 1, characterized in that, The step of smoothing the initial planned trajectory to obtain at least one model output trajectory includes: For each initial planned trajectory, the target sampling points corresponding to the initial planned trajectory are obtained after filtering the initial planned trajectory. Using a pre-built penalty function, the target sampling points corresponding to the initial planned trajectory are optimized to obtain the model output trajectory corresponding to the model output trajectory.
3. The determination method according to claim 1, characterized in that, The step of generating at least one safety fallback planning trajectory based on the acquired local map related to the target vehicle using a homotopy trajectory generation method includes: Obtain a local map containing static and dynamic obstacle information related to the target vehicle; Based on the drivable area segmented by obstacles in the local map, lateral and longitudinal decisions are made. By combining the lateral and vertical decisions, at least one homotopic traversal decision is generated; Based on the homotopy traffic decision, a mathematical optimization algorithm is invoked to generate at least one safe fallback planning trajectory for the target vehicle.
4. The determination method according to claim 1, characterized in that, The step of arbitrarily evaluating the model output trajectory and the safety fallback planned trajectory, and using the evaluated optimal trajectory as the executable trajectory for the current target vehicle, includes: The model output trajectory and the safety fallback planning trajectory are respectively input into the target evaluator to obtain the evaluation score of each model output trajectory and each safety fallback planning trajectory; Based on the evaluation score, determine whether the trajectory meets the execution requirements; If the conditions are met, the trajectory with the lowest evaluation score among the qualified trajectories will be determined as the executable trajectory for the current target vehicle.
5. The determination method according to claim 4, characterized in that, The determination method further includes: If the conditions are not met, the executable trajectory determined at the previous sampling time is combined with the current driving data of the target vehicle to fit the current trajectory, and the fitted trajectory is determined as the current executable trajectory of the target vehicle.
6. A device for determining a driving trajectory, characterized in that, The determining device includes: The acquisition module acquires at least one initial planned trajectory output by the decision planning model for the target vehicle. The smoothing module smooths the initial planned trajectory to obtain at least one model output trajectory. The generation module generates at least one safety fallback planning trajectory for the target vehicle based on the acquired local map related to the target vehicle using the homotopy trajectory generation method. The evaluation module arbitrates and evaluates the model output trajectory and the safety fallback planning trajectory, and uses the optimal trajectory after evaluation as the executable trajectory of the current target vehicle.
7. The determining device according to claim 6, characterized in that, The smoothing module is specifically used for: For each initial planned trajectory, the target sampling points corresponding to the initial planned trajectory are obtained after filtering the initial planned trajectory. Using a pre-built penalty function, the target sampling points corresponding to the initial planned trajectory are optimized to obtain the model output trajectory corresponding to the model output trajectory.
8. The determining device according to claim 6, characterized in that, The generation module is specifically used for: Obtain a local map containing static and dynamic obstacle information related to the target vehicle; Based on the drivable area segmented by obstacles in the local map, lateral and longitudinal decisions are made. By combining the lateral and vertical decisions, at least one homotopic traversal decision is generated; Based on the homotopy traffic decision, a mathematical optimization algorithm is invoked to generate at least one safe fallback planning trajectory for the target vehicle.
9. An electronic device, characterized in that, include: The device includes a processor, a storage medium, and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the determination method as described in any one of claims 1 to 5.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the determination method as described in any one of claims 1 to 5.