Vehicle Trajectory Planning Around Actuator Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle trajectory planning systems face challenges in efficiently incorporating actuator limitations and dynamics, leading to computational inefficiencies and potential deviations from desired trajectories due to windup effects, especially when actuator limitations are unknown or unaccounted for.
Innovation Solution
A method for controlling vehicle trajectories that defines a search space for manipulated variables based on actuator limits, allowing for efficient planning and reduced computational effort by separating actuator dynamics from the planning process, using sensors to detect surroundings and account for road forces and actuator capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If model predictive approaches are used to incorporate actuator limitations and dynamics in planning, then trajectory drivability and reliability are improved, but computational effort increases excessively
Solution Approach 1:
The planning system is segmented into two independent modules: a trajectory planner that generates desired trajectories without complex actuator models, and a controller that handles actuator dynamics and limitations. This segmentation allows each module to operate with simplified computations while the overall system maintains high reliability through the controller's explicit handling of actuator constraints.
Solution Approach 2:
A communication interface is introduced between the trajectory planner and controller to transfer trajectory information and actuator status data. This intermediary mechanism allows the planner to operate independently with low computational overhead while the controller translates trajectories into actuator commands that respect physical limitations, resolving the contradiction between planning simplicity and execution reliability.
2Manufacturing precision
If complex models incorporating actuator dynamics are used in planning, then trajectory accuracy is improved, but real-time computation becomes infeasible
Solution Approach 1:
The system separates trajectory generation from actuator control, allowing the trajectory planner to focus on path accuracy using simplified models while the controller handles the computationally intensive actuator dynamics. This enables real-time operation with high trajectory accuracy without requiring complex models in the planning phase.
Solution Approach 2:
The controller pre-processes trajectory information and pre-calculates actuator commands that account for dynamics and limitations before execution. This preliminary action ensures that trajectory accuracy is maintained while avoiding real-time computational delays during actual vehicle operation.
3Productivity
If actuator limitations are not considered in planning, then computational effort is reduced, but windup effects occur causing trajectory deviations
Solution Approach 1:
The controller acts as an intermediary between the trajectory planner and the actuator, translating desired trajectories into feasible commands that respect actuator limitations. This intermediary layer prevents windup effects by explicitly considering actuator constraints in the control layer without burdening the planning layer with complex computations.
Solution Approach 2:
The system implements feedback from the controller to the trajectory planner, providing information about actuator limitations and achieved trajectory progress. This feedback mechanism allows the planner to adjust future trajectory segments to ensure adherence while maintaining computational efficiency in the planning process.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
The invention relates to a method for steering a vehicle (1) along a trajectory, in which the vehicle (1) comprises a control device (2) which plans the trajectory within a predetermined search range of the trajectory and can engage with the actuators (3, 4, 5) of the vehicle (1) for steering the vehicle (1). At least one limiting value is determined for at least one adjustment variable of an actuator (3, 4, 5) and a search range (9) of the adjustment variable is determined using the at least one limiting value, wherein the search range (9) is taken into account for planning the trajectory.