Occupancy Grid Caching for Faster Autonomous Vehicle Motion Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion planning techniques for autonomous vehicles are computationally intensive and time-consuming, particularly when generating motion plans that require frequent queries about potential intersections with objects in the environment, leading to high resource consumption.
Innovation Solution
The development of an occupancy grid movie system that generates time-stepped occupancy grids for future times, allowing for reduced computational resources and query time by using cached query objects to determine cell occupancy and predict object positions, enabling efficient motion planning and control of the vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional motion planning techniques are used to perform frequent queries about potential intersections with objects in the environment, then motion planning accuracy is improved, but computational resource consumption increases and query time increases
Solution Approach 1:
The system performs preliminary actions by generating an occupancy grid movie that predicts future occupancy states of grid cells ahead of time. Instead of computing intersections on-demand during motion planning queries, the system pre-computes which cells will be occupied by tracked objects at future time steps, storing this information in a cache. This allows motion planning queries to quickly retrieve pre-computed occupancy information without performing expensive geometric intersection calculations in real-time.
Solution Approach 2:
The system creates a simplified copy of the environment state in the form of an occupancy grid movie, which is a temporal sequence of occupancy grids representing predicted future states. This copy contains essential information about object positions and potential intersections without requiring access to the full complex sensor data and object models. The motion planning system queries this simplified representation rather than performing detailed geometric checks against original sensor data.
2Reliability
If the number of motion planning queries is increased to improve path safety, then navigation reliability is improved, but the time and computational resources required for each query increase
Solution Approach 1:
The system performs preliminary actions by generating an occupancy grid movie that predicts future occupancy states of grid cells ahead of time. Instead of computing intersections on-demand during motion planning queries, the system pre-computes which cells will be occupied by tracked objects at future time steps, storing this information in a cache. This allows motion planning queries to quickly retrieve pre-computed occupancy information without performing expensive geometric intersection calculations in real-time.
Solution Approach 2:
The system implements dynamics by creating a temporal model of the environment through the occupancy grid movie, which captures the dynamic movement of tracked objects over time. The movie represents a sequence of occupancy grids at different future time steps, allowing the motion planning system to query not just current but also future occupancy states. This dynamic representation enables the system to perform multiple safety checks across different time points without re-computing from scratch each time.
Data Source
AI summary
Various technologies described herein pertain to generating an occupancy grid movie for utilization in motion planning for the autonomous vehicle. The occupancy grid movie can be generated for a given time and can include time-stepped occupancy grids for future times that are at predefined time intervals from the given time. The time-stepped occupancy grids include cells corresponding to regions in an environment surrounding the autonomous vehicle. Probabilities can be assigned to the cells specifying likelihoods that the regions corresponding to the cells are occupied at the future times. Moreover, cached query objects that respectively specify indices of cells of a grid occupied by a representation of an autonomous vehicle at corresponding orientations are described herein. An occupancy grid for the environment surrounding the autonomous vehicle can be queried to determine whether cells of the occupancy grid are occupied utilizing a cached query object from the cache query objects.