Object Trajectory Relevance Filtering for Vehicle Path Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Perception and prediction systems in vehicles require significant computing resources and time to determine actions due to the large amount of data processing needed to handle numerous objects in the environment, limiting vehicle performance.
Innovation Solution
A relevancy filter system is used to determine relevant predicted object trajectories for candidate vehicle trajectories, reducing unnecessary data storage and processing by assigning relevancy scores based on predicted object trajectories' relevance to candidate vehicle trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If perception systems process all sensor data from numerous objects in the environment, then object detection completeness is improved, but computing resource consumption increases
Solution Approach 1:
The system extracts only the relevant subset of object trajectories that are pertinent to candidate vehicle trajectories, filtering out unnecessary data. This is achieved by determining relevancy scores for individual predicted object trajectories based on their relationship to candidate vehicle trajectories, and selectively processing only those with high relevancy scores.
Solution Approach 2:
The system applies different processing quality levels to different objects based on their relevancy. Objects with high relevancy scores to candidate vehicle trajectories receive detailed processing, while objects with low relevancy scores are filtered out or processed minimally, creating a localized quality approach to data processing.
2Reliability
If perception systems process all sensor data from numerous objects in the environment, then object detection completeness is improved, but processing time increases
Solution Approach 1:
The system performs preliminary filtering by determining relevancy scores for predicted object trajectories before full processing. This preliminary action identifies and separates relevant objects from irrelevant ones early in the processing pipeline, preventing unnecessary processing time from being spent on objects that will not impact vehicle trajectory decisions.
3Reliability
If planning systems evaluate all predicted object trajectories for candidate vehicle trajectories, then navigation safety is improved, but data processing overhead increases
Solution Approach 1:
The system extracts and processes only the relevant predicted object trajectories that have high relevancy scores to candidate vehicle trajectories. By filtering out irrelevant trajectories using relevancy scoring, the system reduces data processing overhead while maintaining navigation safety through focused evaluation of critical objects.
Solution Approach 2:
The system performs partial processing by evaluating only a subset of predicted object trajectories that are deemed relevant based on relevancy scores. This partial action approach processes sufficient data to ensure navigation safety without the excessive processing overhead of evaluating all possible object trajectories.
Data Source
AI summary
Systems and techniques for whether to associate predicted trajectories of objects detected in an environment with candidate vehicle trajectories are described. A vehicle traversing an environment may have several candidate actions that may be used to control the vehicle. Predicted object trajectories may be excluded from further processing based on a relevancy score determination. Various factors may be evaluated to determine the relevancy score associated with a predicted object trajectory and candidate vehicle action pair that indicated the impact of the predicted object trajectory and candidate vehicle action. A resultant trajectory may be determined using only those relevant predicted object trajectories.