Curvilinear Motion Planning for Autonomous Vehicle Trajectory Conflicts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicle motion planning is computationally challenging due to high-dimensional state spaces and the need for dynamic, safe, rule-compliant, and comfortable trajectory planning, especially when dealing with moving road users and ambiguous transformations in curvilinear coordinates.
Innovation Solution
A motion planning system determines a reference curve and segments the time horizon into intervals to predict lateral offset distances, using these to identify occupied and free configuration spaces, and adjusts trajectories to avoid conflicts, employing coarse-to-fine bounding geometry determination and lookup tables for efficient computation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If motion planning is performed in curvilinear coordinates to reduce dimensionality, then computational complexity is reduced, but mathematical ambiguity arises near the center of curvature of reference curves
Solution Approach 1:
The patent segments the time horizon into multiple time intervals and further segments each time interval into subintervals. This segmentation allows the system to handle the curvilinear coordinate transformation in discrete steps, reducing the computational burden while maintaining position accuracy by updating coordinates at each segment boundary rather than continuously.
Solution Approach 2:
The patent performs preliminary action by pre-calculating and storing curvilinear coordinates for multiple points along the reference curve before motion planning begins. This pre-computation creates a lookup table that eliminates the need for real-time coordinate transformation during planning, resolving the mathematical ambiguity issue while maintaining reduced computational complexity.
2Measurement precision
If distances to moving actors are re-computed frequently to maintain accuracy, then measurement precision is improved, but computational resources are consumed
Solution Approach 1:
The patent merges the distance computation with the coordinate transformation by calculating curvilinear coordinates for multiple points along the reference curve simultaneously. This combined approach computes distances to moving actors as part of the coordinate transformation process rather than as separate operations, improving measurement precision while reducing overall computational resource consumption.
3Adaptability or versatility
If the reference curve is updated frequently to adapt to lane changes, then adaptability is improved, but computational overhead increases
Solution Approach 1:
The patent implements dynamics by allowing the reference curve to be updated dynamically when lane changes are detected, while using pre-computed coordinate tables to minimize the computational overhead of these updates. The system adapts to new lanes by selecting appropriate pre-computed coordinate sets rather than recalculating everything, maintaining adaptability while controlling computational complexity.
Data Source
AI summary
Methods for planning a trajectory for an autonomous vehicle are disclosed. A vehicle motion planning system will determine a reference curve that represents a path via which the vehicle may travel. The system will detect an actor that is moving in the environment. The system will segment the reference curve according to time intervals. For each of the time intervals, the system will identify a bounding geometry for the actor, predict a lateral offset distance between the reference curve and the actor, and use the predicted lateral offset distance to determine whether to alter a planned trajectory for the autonomous vehicle.