Autonomous Driving Sensor Fusion With Tracking Feedback Loops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous driving systems lack effective feedback mechanisms that allow earlier processing units to improve their performance based on feedback from later units, leading to suboptimal object tracking and positioning across multiple views from various sensors.
Innovation Solution
The implementation of a feedback system in the processing loop of autonomous driving systems, where units such as the perception unit, feature fusion unit, and object tracking unit can provide feedback to earlier units, enhancing the accuracy and reliability of object detection and positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is passed linearly through processing units without feedback, then the system structure is simple, but object tracking accuracy deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where later processing units (tracking unit, prediction unit, planning unit) provide feedback information back to earlier units (perception unit, feature fusion unit). This feedback enables earlier units to refine their processing based on outcomes from downstream units, thereby improving object tracking accuracy while maintaining a manageable system structure through organized feedback loops.
2Measurement precision
If feedback loops are added to improve tracking, then object detection accuracy improves, but processing time increases
Solution Approach 1:
The patent applies preliminary action by performing feature fusion and object detection in advance before final tracking and planning. The perception unit and feature fusion unit prepare processed data beforehand, creating a foundation that reduces the computational burden during feedback iterations. This preliminary processing helps mitigate the time increase that would otherwise result from multiple feedback loops.
Data Source
AI summary
A sensor data processing system includes various elements, including a perception unit that collects data representing positions of sensors on a vehicle and obtains environmental information around the vehicle via the sensors. The sensor data processing system also includes a feature fusion unit that combines the first environmental information from the sensors into first fused feature data representing first positions of objects around the vehicle, provides the first fused feature data to the object tracking unit, receives feedback for the first fused feature data from the object tracking unit, and combines second environmental information from the sensors using the feedback into second fused feature data representing second positions of objects around the vehicle. The sensor data processing system may then at least partially control operation of the vehicle using the second fused feature data.