Track-Based Object Association for Low-Latency Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Processing large volumes of transportation network data from various sensors to provide meaningful insights into road user behavior is complicated by latency and inconsistencies, making it difficult to associate and fuse data from distributed sensing sources effectively.
Innovation Solution
Implementing a track-based moving object association method that uses processors to receive and process position information from multiple sensors, determine object tracks, calculate similarity measures, and merge matched objects into a single tracked object track, while outputting unmatched objects as separate entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data from multiple distributed sensors is collected and processed to provide insights into road user behavior, then the quantity and coverage of sensing data increases, but processing latency and computational complexity increase
Solution Approach 1:
The patent segments the processing of sensor data by creating separate tracking pipelines for each sensor source. Each sensor's position information is processed independently to determine object tracks, and then these tracks are associated through similarity measures. This segmentation allows parallel processing of multiple sensor streams, reducing overall processing latency while maintaining comprehensive data coverage.
2Measurement precision
If position information from multiple sensors is processed to determine object tracks and associations, then the accuracy of object association improves, but computational complexity and processing difficulty increase
Solution Approach 1:
The patent applies preliminary action by determining object tracks for each sensor independently before performing the complex association operation. Each sensor's position information is first processed to establish individual tracks, which then serve as inputs to the association algorithm. This preliminary tracking simplifies the subsequent association step, as it only needs to match pre-established tracks rather than processing raw position data from multiple sensors simultaneously.
Solution Approach 2:
The patent introduces similarity measures as an intermediary mechanism to bridge tracks from different sensors. Instead of directly comparing raw position information from multiple sensors, the system computes similarity measures between established tracks, which serves as an intermediate representation that simplifies the association process while maintaining high accuracy in matching objects across sensor sources.
3Reliability
If tracks from multiple sensors are associated using similarity measures, then the reliability of tracked object identification improves, but processing time increases
Solution Approach 1:
The patent implements partial action by computing similarity measures selectively for track pairs that are likely to be associated, rather than exhaustively comparing all possible track combinations. The system processes position information from multiple sensors to determine tracks, then applies similarity measures to associate these tracks, achieving reliable object identification while avoiding unnecessary computational overhead from exhaustive comparisons.
Data Source
AI summary
A track based moving object association is described for distributed sensing applications, such as for identifying multiple objects within a vehicle transportation network for use by a vehicle navigating the network. Position information for the multiple objects within a portion of the vehicle transportation network is obtained from multiple sensors within the portion of the vehicle transportation network. Using the position information of respective sensors of the multiple sensors, a respective track for objects of the multiple objects is determined. Similarity measures are determined for multiple tracks, including at least a first track determined using the position information of a first sensor of the multiple sensors and a second track determined using the position information of a second sensor of the multiple sensors. Based on the similarity measures of the tracks, a tracked object track for a tracked object of the multiple objects is determined.