Dynamic Vehicle Representation for Accurate Collision Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle planning systems in autonomous and semi-autonomous vehicles rely on fixed buffers to predict collisions, which can lead to overly cautious behavior, preventing vehicles from proceeding past objects when there is sufficient space, thus limiting progress and efficiency.
Innovation Solution
A vehicle safety system that dynamically determines a vehicle representation by adjusting a buffer region based on historical and sensor data, including position and heading errors, to accurately predict potential collisions and optimize vehicle actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed buffers are used to predict collisions, then safety is improved, but navigation efficiency deteriorates
Solution Approach 1:
The patent applies dynamics by transforming the static fixed buffer into a dynamic vehicle representation that adapts its size and shape based on real-time vehicle state data. The buffer region is continuously adjusted according to actual position and heading errors, allowing the system to maintain accurate collision prediction while adapting to changing vehicle conditions, thereby resolving the contradiction between safety and navigation efficiency
Solution Approach 2:
The patent implements parameter changes by modifying the buffer region parameters (size, shape, orientation) based on measured vehicle state deviations. The buffer dimensions are dynamically scaled according to position error and heading error parameters, enabling the system to expand the buffer when uncertainty is high and shrink it when the vehicle tracks the trajectory accurately, thus balancing safety and efficiency
2Reliability
If fixed buffers are used to predict collisions, then safety is improved, but vehicle progress is limited
Solution Approach 1:
The patent uses parameter changes to dynamically adjust the buffer region size based on actual vehicle performance. When the vehicle maintains accurate trajectory tracking, the buffer is reduced, allowing the vehicle to proceed without unnecessary conservative actions. When deviations occur, the buffer expands to ensure safety, thus eliminating unnecessary braking or swerving while maintaining collision avoidance
Solution Approach 2:
The patent implements feedback by continuously measuring actual vehicle position and heading against the planned trajectory, then using these error measurements to adjust the buffer region. This closed-loop feedback mechanism allows the system to learn from actual vehicle behavior and adjust collision predictions accordingly, preventing unnecessary safety actions while maintaining protection when needed
Data Source
AI summary
Techniques for accurately predicting vehicle state errors for avoiding collisions with objects detected in an environment of a vehicle are discussed herein. A vehicle safety system can implement a model determine a representation of the vehicle usable in a scenario. The model may dynamically determine a size and/or a heading of the vehicle representation based on differences between a candidate trajectory and a current trajectory of the vehicle. The safety system can identify a potential collision between the vehicle and the object based at least in part on an overlap of the vehicle representation and an object representation.