Dual-Threshold Object Recognition for Radar Anti-Clutter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing on-vehicle object recognition systems face challenges in achieving both a large recognition region and high anti-clutter characteristics, as setting a high threshold to eliminate clutter noise can result in lost object information in far distance regions, while setting a low threshold may not adequately remove clutter, especially when using radar sensors like LIDAR.
Innovation Solution
The system employs a dual-threshold approach, using a low first threshold for RFS tracking in far distance regions to maintain high anti-clutter characteristics and a high second threshold for conventional tracking in near distance regions to reduce processing load and prevent erroneous recognition, allowing for extended recognition regions without excessive processing overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a high threshold is set to eliminate clutter noise, then anti-clutter characteristics are improved, but object information in far distance regions is lost
Solution Approach 1:
The patent divides the detection region into multiple regions (first region with far distance and second region with near distance) and applies different threshold values to each region. The positional information acquiring unit acquires first observation information using a first threshold for the first region, and second observation information using a second threshold for the second region, thereby preventing information loss in far distance while maintaining anti-clutter performance.
Solution Approach 2:
Different threshold values are applied to different spatial regions based on their specific characteristics. The first threshold (lower) is applied to far distance regions where signal strength is weak, while the second threshold (higher) is applied to near distance regions where clutter is more prominent, optimizing detection performance for each local area.
2Area of stationary object
If a low threshold is set to maintain recognition in far distance regions, then recognition range is extended, but clutter noise is not adequately removed
Solution Approach 1:
The detection space is segmented into multiple regions with different threshold requirements. By separating far distance region detection (using first threshold) from near distance region detection (using second threshold), the system can extend recognition range without allowing clutter noise to dominate in near regions.
Solution Approach 2:
The threshold value is adjusted locally according to the distance region. In far distance regions, a lower threshold maintains sensitivity to weak signals, while in near distance regions, a higher threshold effectively suppresses clutter noise, optimizing detection quality in each local area.
3Productivity
If conventional tracking is used in all regions, then processing load is reduced, but anti-clutter characteristics deteriorate in far distance regions
Solution Approach 1:
The tracking process is segmented into two paths: RFS tracking for the first region (far distance) and conventional tracking for the second region (near distance). This segmentation allows the system to apply computationally intensive RFS tracking only where needed (far distance with clutter issues) while using lighter conventional tracking in near regions.
Solution Approach 2:
The tracking method parameter is changed based on the detection region. RFS (Random Finite Set) tracking, which has better anti-clutter characteristics, is applied to far distance regions, while conventional tracking is used for near distance regions, optimizing the balance between processing load and detection reliability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively balances recognition range and anti-clutter performance, enabling accurate object tracking over a wide area while minimizing processing load and erroneous recognition, particularly by using RFS tracking in far distance regions and conventional methods in near distance regions.
Implementation Method 1
the object sensor emitting probing waves as electromagnetic waves and receiving reflection waves of the probing waves which are reflected at the object
Data Source
AI summary
A positional information acquiring unit that acquires positional information in accordance with an object detection signal of the object sensor includes a first acquiring unit that acquires first observation information as positional information using a first threshold, and a second acquiring unit that acquires second observation information as positional information using a second threshold different from the first threshold. The object tracking unit is provided with a first tracking processing unit that executes a tracking process in accordance with the first observation information and a second tracking processing unit that executes the tracking process in accordance with the second observation information. Either one of the first tracking processing unit or the second tracking processing unit has anti-clutter characteristics higher than that of the other one. A state identifying unit identifies the state of the object based on a result of the tracking process executed by the tracking processing unit.