LiDAR Object Tracking with Group Density Clustering Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of clusters formed by clustering point clouds acquired through LiDAR sensors is deteriorated, posing a challenge in autonomous vehicle navigation.
Innovation Solution
A tuning parameter determination method for LiDAR-based object tracking, involving initial parameter setting, misrecognition storage, and post-processing, along with group density-based clustering techniques to enhance cluster accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If density-based clustering is performed on point cloud data from LiDAR sensor, then object tracking capability is enabled, but cluster accuracy is deteriorated
Solution Approach 1:
The patent applies preliminary action by performing pre-processing steps before clustering, including converting point cloud data into group cloud data where multiple points are aggregated into representative groups. This preliminary aggregation improves the quality of input data for subsequent density-based clustering, thereby enhancing cluster accuracy while maintaining automation. The pre-processing includes calculating representative positions and filtering duplicate groups, which prepares the data structure for more accurate clustering execution.
2Measurement precision
If tuning parameter is optimized for clustering accuracy, then misrecognition situations are reduced, but system complexity increases due to database storage and post-processing requirements
Solution Approach 1:
The patent implements feedback by storing misrecognition situations in a database and using this stored information to detect and adjust tuning parameters. The system performs post-processing on detected parameters and feeds back improved parameters to the clustering process. This closed-loop feedback mechanism reduces misrecognition situations and improves clustering accuracy while managing system complexity through structured parameter optimization.
3Productivity
If group density-based clustering is performed on point cloud, then processing efficiency is improved, but cluster accuracy may be compromised due to data conversion
Solution Approach 1:
The patent applies segmentation by dividing the point cloud data into groups before clustering, where multiple points are aggregated into representative group entities. This segmentation into group cloud data improves processing efficiency by reducing the number of individual points to be processed. The method maintains accuracy by calculating representative positions for each group and preserving spatial relationships, thereby achieving efficient processing without compromising cluster accuracy.
Data Source
AI summary
An object-tracking apparatus using a LiDAR sensor of an embodiment includes: a LiDAR sensor configured to acquire a point cloud related to an object; a first clustering unit configured to primarily cluster the point cloud; and a second clustering unit configured to secondarily cluster the result of primary clustering in response to the time taken to perform primary clustering.