LiDAR Point Clustering with Noise Separation for Object Tracking

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

LiDAR sensors in autonomous vehicles face accuracy issues due to noise from floating particles in the air and sensor-generated noise, affecting object recognition.

Innovation Solution

A method and apparatus for tracking objects using LiDAR sensors that determine noise points based on spacing distances between target and neighboring points, clustering valid points to enhance object tracking performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If LiDAR sensor is used to acquire points for object tracking, then object tracking capability is enabled, but noise points from floating particles and sensor noise deteriorate recognition accuracy

Engineering Contradiction:

Improveobject tracking capabilityVSAvoidrecognition accuracy

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The patent extracts and removes noise points from the point cloud data by analyzing spacing distances between adjacent points. Points with spacing distances greater than a threshold are identified as noise and excluded from clustering, thereby improving recognition accuracy while preserving valid object tracking capability

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary noise determination process between point acquisition and clustering. This intermediary step uses spacing distance analysis to filter noise points before they affect the clustering results, acting as a mediator that protects the final object recognition from noise interference

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If all acquired points are used for clustering, then processing is simplified, but noise points reduce tracking accuracy

Engineering Contradiction:

Improveprocessing complexityVSAvoidtracking accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary noise determination before clustering by analyzing spacing distances between adjacent points in advance. This preliminary action identifies and flags noise points so they can be excluded during the subsequent clustering process, improving tracking accuracy without significantly increasing overall processing complexity

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If noise determination region size is increased to capture more floating particle noise, then noise detection capability improves, but processing time increases

Engineering Contradiction:

Improvenoise detection capabilityVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies partial action by determining noise information only for points within a specific noise determination region rather than processing the entire point cloud. This selective approach captures sufficient floating particle noise while reducing the overall processing time and computational load

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12546893B2Method and apparatus for tracking object using LiDAR sensor and recording medium storing program to execute the method

Publication Date: 2026.02.10 HYUNDAI MOTOR CO LTD

AI summary

A method of tracking an object using a LiDAR sensor includes determining, by an information determiner, based on the spacing distance between a target point and a valid neighboring point disposed adjacent to the target point in the same layer as the target point, among first points acquired by the LiDAR sensor, noise information indicating whether the target point is a noise point; and clustering, by a clustering unit, the first points using the noise information.