Object Recognition Fusion for Autonomous Driving Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional object recognition processing apparatuses generate multiple tracks for a single object due to differences in accuracy between sensors, leading to inaccurate distance and position recognition between a vehicle and peripheral objects.
Innovation Solution
An object recognition processing apparatus that uses multiple object detection units to fuse detection information, select candidate combinations of object fusion signals, and determine whether these combinations represent the same object, thereby integrating signals for accurate recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If sensor fusion technology uses multiple sensors with different accuracies to detect peripheral objects, then the coverage of object detection is improved, but multiple tracks are generated for the same object leading to recognition errors
Solution Approach 1:
The patent merges multiple detection results from different sensors by determining whether detection information from sensors with different accuracies corresponds to the same object. When matches are found, the system combines the detection results into a single unified track, eliminating duplicate tracks while preserving the benefits of multi-sensor detection coverage.
2Reliability
If multiple sensors with different accuracies are used simultaneously, then sensing performance is enhanced, but the complexity of data processing increases
Solution Approach 1:
The patent segments the complex sensor fusion process into distinct processing steps: first determining whether detection information from different sensors corresponds to the same object, then selectively combining results only when matches are found. This segmentation reduces processing complexity by avoiding unnecessary computations on non-matching detection data.
Solution Approach 2:
The system performs preliminary determination of object correspondence before combining detection results. By pre-identifying which detection information corresponds to the same object, the system avoids complex real-time processing of all possible sensor combinations, thereby reducing overall data processing complexity.
Data Source
AI summary
An object recognition processing apparatus includes: an object fusion processing unit 6 for fusing detection information signals of objects existing around a vehicle, which are detected by a plurality of object detection units to output object fusion signals; a same object fusion result combination candidate selection processing unit for selecting, for a combination of object fusion signals, candidates of the object fusion signals to be combined; a same object fusion result combination determination processing unit for determining whether or not a combination of the selected object fusion signals is a combination of object fusion signals of the same object; and a same object fusion combination integration processing unit for recognizing the objects existing around the vehicle on the basis of a result of the determination.