Vehicle Filter Tracking Bounding Shape Vertices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately tracking the location and motion of perceived vehicles due to incomplete sensor observations, which can lead to inaccurate motion planning and increased collision risks.
Innovation Solution
A vehicle filter system employing a motion model and an observation model to predict and correct the location of vehicle bounding shapes using dominant vertices and sides, rather than relying solely on the centroid or footprint of observation bounding shapes, thereby improving tracking accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor observations are used to track vehicle locations, then the autonomous vehicle can perceive its environment, but incomplete sensor observations lead to inaccurate tracking
Solution Approach 1:
The system performs preliminary prediction of vehicle bounding shape locations using a motion model before receiving sensor observations. This allows the system to have an anticipated state ready, which is then refined when observations arrive, preventing information loss rather than just compensating for it
Solution Approach 2:
The system implements a feedback loop where sensor observations are used to correct predicted bounding shape locations. The correction process feeds back into the tracking system to improve future predictions, continuously reducing the impact of incomplete observations on tracking accuracy
2Measurement precision
If the centroid or footprint of observation bounding shapes is used for tracking, then the system is simple to implement, but tracking accuracy is reduced
Solution Approach 1:
The system segments the vehicle bounding shape into multiple vertices and sides, tracking each component separately rather than using a single centroid point. This segmentation allows for more precise tracking of the vehicle's actual geometry and motion while maintaining manageable system complexity through modular processing
Solution Approach 2:
The system applies different tracking approaches to different parts of the bounding shape (vertices and sides have different properties than the centroid). Each vertex and side can be tracked with appropriate precision based on its local characteristics, improving overall tracking accuracy without uniformly increasing complexity across the entire system
Data Source
AI summary
The present disclosure provides systems and methods for tracking vehicles or other objects that are perceived by an autonomous vehicle. A vehicle filter can employ a motion model that models the location of the tracked vehicle using a vehicle bounding shape and an observation model that generates an observation bounding shape from sensor observations. A dominant vertex or side from each respective bounding shape can be identified and used to update or otherwise correct one or more predicted shape locations associated with the vehicle bounding shape based on a shape location associated with the observation bounding shape.