Autonomous Vehicle Trajectory Planning With Joint Lateral-Speed Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional trajectory planning methods for autonomous driving vehicles are inadequate in terms of robustness and adaptability to complex road conditions and environments, as they primarily rely on lane line constraints and fail to effectively combine lateral displacement and longitudinal speed planning.
Innovation Solution
A trajectory planning method that integrates lateral displacement and longitudinal speed planning by receiving perception data, positioning data, and map information to generate a preliminary traveling trajectory, which is then refined to produce a target traveling trajectory with predicted drivable speeds at each road point, considering traffic road conditions and allowable lateral errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If trajectory planning is performed only based on lane line constraint, then the vehicle can travel without crossing lines, but the planning result is poor in robustness and cannot be adapted to complex road conditions
Solution Approach 1:
The patent merges lane line constraints with virtual constraints to form a comprehensive constraint system. The virtual constraints are generated based on road conditions and environment, and are combined with lane line constraints to provide more robust and adaptable trajectory planning for complex scenarios.
Solution Approach 2:
The patent introduces dynamic virtual constraints that can adapt to changing road conditions and environments. These virtual constraints are not fixed but are dynamically generated and adjusted based on the current situation, allowing the trajectory planning to be both robust and adaptable.
2Manufacturing precision
If lateral displacement planning and longitudinal speed planning are performed separately, then the planning process is simpler, but the safety precision and comfort of the planned trajectory are reduced
Solution Approach 1:
The patent combines lateral displacement planning and longitudinal speed planning into a unified joint planning process. By considering both lateral and longitudinal constraints simultaneously, the system achieves higher safety precision and comfort while managing complexity through integrated constraint handling.
Data Source
AI summary
A trajectory planning method for an autonomous vehicle includes a lateral displacement planner that plans a preliminary traveling trajectory for an autonomous driving vehicle based on perceptual data, positioning data, and map information, where the preliminary traveling trajectory includes N road points, and each road point includes coordinate information of the road point and allowable lateral error information of the road point. A longitudinal speed planner inherits all or some of the N road points output by the lateral displacement planner and determines speed information of a subset of the N road points based on traffic road condition information and the perception data to obtain a target traveling trajectory. The speed information is a drivable speed of the autonomous driving vehicle. The target traveling trajectory includes the subset of the N road points, each including the drivable speed of the autonomous driving vehicle.