Kurtosis-Based Track Pruning in Sensor Fusion Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sensor-fusion systems face computational challenges in tracking multiple classes of objects simultaneously due to combinatorial explosions, which can lead to processing lag and inefficiencies, especially in complex environments with various sensors like radar, cameras, and lidar.
Innovation Solution
The implementation of Kurtosis-based pruning, which involves determining weights for candidate object-tracks from different sensors, calculating the distribution of these weights, and pruning less likely matches based on Kurtosis values to reduce the number of comparisons needed, thereby preventing combinatorial explosions and enhancing processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor-fusion systems track multiple classes of objects simultaneously using multiple sensors, then tracking accuracy and safety are improved, but computational complexity increases leading to combinatorial explosion
Solution Approach 1:
The patent segments the tracking process into distinct phases: generating candidate tracks from individual sensors, applying Kurtosis-based pruning to filter candidates, and performing final matching only on pruned sets. This segmentation breaks the combinatorial explosion into manageable stages, allowing accurate multi-class tracking while controlling computational complexity at each step.
Solution Approach 2:
The patent applies preliminary action by performing Kurtosis-based pruning before the final matching stage. Candidate tracks are pre-filtered using statistical analysis of weight distributions, eliminating unlikely matches in advance. This preliminary filtering reduces the computational burden of subsequent matching operations while preserving tracking accuracy.
2Measurement precision
If sensor-fusion systems perform exhaustive matching of candidate tracks from multiple sensors, then matching accuracy is improved, but processing speed decreases due to combinatorial explosion
Solution Approach 1:
The patent changes the parameter space by introducing Kurtosis values as a new filtering criterion. Instead of directly matching all candidate pairs, the system transforms candidate tracks into a pruned set based on Kurtosis analysis of weight distributions. This parameter transformation enables accurate matching by filtering out implausible candidates before the matching process, improving processing speed without sacrificing accuracy.
Solution Approach 2:
The patent extracts and removes unlikely candidate tracks from the matching process through Kurtosis-based pruning. By identifying and eliminating candidates with low probability of being true matches, the system reduces the number of comparisons needed while maintaining high matching accuracy for the remaining candidates.
3Productivity
If sensor-fusion systems specialize in tracking certain object types, then processing speed is improved, but tracking versatility deteriorates
Solution Approach 1:
The patent implements universality by designing a Kurtosis-based pruning framework that works across multiple sensor types (radar, camera, lidar) and multiple object classes (vehicles, pedestrians, cyclists). The statistical pruning mechanism is class-agnostic and sensor-agnostic, allowing the system to maintain high processing speed while tracking diverse object types simultaneously, thus achieving both speed and versatility.
Data Source
AI summary
Kurtosis based pruning for a sensor-fusion system includes aligning candidate radar tracks with one of multiple candidate visual tracks. For each candidate vision track, a weight or other evidence of matching is assigned to each candidate radar track. An inverse of matching errors between each candidate vision track and each candidate radar track contributes to the evidence, which is normalized to produce, for each candidate vision track, a distribution associated with all candidate radar tracks. A Kurtosis or a shape of this distribution is calculated. Based on Kurtosis values, some candidate radar tracks are selected for matching and other remaining candidate radar tracks are pruned. The Kurtosis is used to determine how many candidates to retain and how many to prune.