Sensor Track Association Using Occlusion Constraints in Vehicle Perception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing perception systems in vehicles face challenges in efficiently and accurately processing large sets of sensor data from multiple types of sensors, leading to potential delays and inaccuracies in object tracking, which can result in collisions.
Innovation Solution
The use of occlusion constraints to modify radar track data and vision track data, de-emphasizing pairs of tracks that appear to be occluded, thereby improving the speed and accuracy of object identification within a field of view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sophisticated processing hardware is used to handle conflicting sensor data, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system performs preliminary occlusion analysis on sensor data before full processing. By pre-identifying occluded tracks and filtering them out early in the processing pipeline, the system reduces the computational burden on subsequent processing stages, thereby reducing hardware complexity requirements while maintaining tracking accuracy through selective processing of only relevant data.
2Measurement precision
If all sensor data is processed to ensure accurate object tracking, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system extracts and removes occluded tracks from the sensor data set before processing. By identifying and eliminating data points that are occluded (using occlusion constraints based on spatial relationships and occlusion probabilities), the system processes only the relevant, non-occluded data, thereby reducing processing time while maintaining tracking accuracy for visible objects.
3Productivity
If occlusion constraints are applied to filter track associations, then productivity is improved, but measurement precision may worsen
Solution Approach 1:
The system dynamically adjusts occlusion probability thresholds and association costs based on environmental conditions and sensor reliability. By changing these parameters adaptively, the system can filter out false associations efficiently (improving productivity) while maintaining accurate tracking of valid objects. The parameter adjustments ensure that genuine tracks are not incorrectly filtered while occluded tracks are effectively removed.
Data Source
AI summary
Perception systems and methods include use of occlusion constraints for resolving tracks from multiple types of sensors. An occlusion constraint is applied to an association between a radar track and vision track to indicate a probability of occlusion. The perception systems and methods refrain from evaluating occluded and collected radar tracks and vision tracks. The probability of occlusion is utilized for deemphasizing pairs of radar tracks and vision tracks with a high likelihood of occlusion and therefore, not useful for tracking. Improved perception data is provided and more closely represents multiple complex data sets for a vehicle to prevent a collision with an occluded object as the vehicle operates in an environment.