Hierarchical Collision Grids for Fast Trajectory Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles lack efficient methods to evaluate planned trajectories for potential collisions with surrounding objects, which can lead to unsafe driving scenarios.
Innovation Solution
A method and system using a set of grids with adjustable cell sizes based on object dimensions to evaluate planned trajectories for potential collisions, allowing for efficient collision detection and trajectory optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional collision detection methods are used, then collision detection can be performed, but the detection time is excessive and efficiency is low
Solution Approach 1:
The patent divides the spatial environment into multiple grid layers with different resolutions. A first grid layer with coarse resolution is used for initial collision detection, and a second grid layer with fine resolution is used for detailed detection only in relevant regions. This segmentation allows the system to avoid processing all areas at high resolution, significantly reducing computation time while maintaining detection accuracy.
2Measurement precision
If high-resolution grid is used for all areas, then collision detection precision is improved, but computational complexity and processing time increase
Solution Approach 1:
The patent applies different grid resolutions to different spatial regions based on their importance. High-resolution grids are applied only to regions near the autonomous vehicle or regions with high collision risk, while low-resolution grids are used for distant or less critical areas. This local quality approach ensures high detection precision where needed while reducing overall computational complexity.
Solution Approach 2:
The patent introduces a hierarchical dimension to the grid structure, organizing grids into multiple layers with different resolutions. This dimensional hierarchy allows the system to navigate between coarse and fine levels of detail, processing only necessary regions at high resolution and thereby reducing computational complexity while maintaining precision where required.
3Reliability
If comprehensive trajectory validation is performed, then collision avoidance reliability is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary collision detection using a coarse-resolution first grid layer before finalizing the trajectory. This preliminary action identifies obvious collision risks early in the planning process, allowing the system to reject unsafe trajectories quickly without performing time-consuming fine-resolution analysis on obviously invalid paths, thus improving reliability while reducing overall processing time.
Solution Approach 2:
The patent applies fine-resolution grid analysis only to specific regions where collision risk is detected or where the vehicle will be located, rather than performing comprehensive high-resolution validation across the entire trajectory. This partial action approach maintains reliability for critical regions while significantly reducing processing time by avoiding unnecessary detailed analysis in safe regions.
Data Source
AI summary
Aspects of the disclosure provide for evaluation of a planned trajectory for an autonomous vehicle. For instance, for each of a plurality of objects, a predicted trajectory may be received. The planned trajectory may identify locations and times that the vehicle will be at those locations. For each of the plurality of objects, a grid including a plurality of cells may be generated. Occupancy of each grid for each of the plurality of objects may be determined based on the predicted trajectories. A cell of each grid which will be occupied by the vehicle at a location and time of the planned trajectory may be identified. The planned trajectory may be evaluated based on whether any identified cell is occupied by any of the plurality of objects at the time.