LiDAR Cluster Merging Using Motion Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current LiDAR sensor cluster merging algorithms often lead to over-segmentation or under-segmentation, particularly when the sensor is occluded, as they primarily rely on distance and geometry without considering motion characteristics, resulting in inaccurate merging of clusters representing the same object.
Innovation Solution
A system and method that determine cluster merging by evaluating both the distance between clusters and their motion characteristics, using additional point clouds to calculate alignment errors and determine if clusters are moving at similar speeds, thereby deciding whether to merge them based on predefined thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If cluster merging is based only on distance and geometry, then the merging process is simple and fast, but over-segmentation occurs particularly when the LiDAR sensor is occluded
Solution Approach 1:
The patent changes the parameters used for cluster merging from only geometric parameters (distance, shape) to include motion parameters (velocity, acceleration). By incorporating temporal motion information alongside spatial geometry, the system resolves the contradiction by maintaining simple distance-based filtering while adding motion characteristic comparison to prevent over-segmentation during occlusion events.
Solution Approach 2:
The patent introduces dynamic motion characteristics (velocity and acceleration) into the static cluster merging process. By evaluating how clusters move over time in addition to their spatial relationship, the system dynamically adjusts merging decisions to prevent over-segmentation while maintaining computational efficiency through threshold-based motion comparison.
2Reliability
If additional point clouds are received and alignment operations are performed to determine motion characteristics, then segmentation accuracy improves by reducing over-segmentation, but the processing complexity and time increase
Solution Approach 1:
The patent segments the cluster merging process into two distinct phases: a first phase that performs quick distance-based filtering to eliminate obviously separate clusters, and a second phase that performs computationally intensive motion characteristic analysis only on candidate clusters that pass the distance threshold. This segmentation reduces overall processing complexity while maintaining high segmentation accuracy.
Solution Approach 2:
The patent applies partial action by performing alignment operations and motion characteristic determination only on clusters that satisfy the distance threshold condition, rather than performing these expensive operations on all possible cluster pairs. This selective approach significantly reduces processing complexity while maintaining segmentation accuracy for relevant clusters.
3Measurement precision
If clusters separated by larger distances are merged based on motion characteristics, then object detection accuracy improves, but the risk of merging unrelated objects increases
Solution Approach 1:
The patent merges multiple criteria (distance threshold, velocity similarity, acceleration similarity) into a unified cluster merging decision. By combining spatial proximity with temporal motion consistency, the system enables merging of clusters separated by larger distances while preventing false merging of unrelated objects that would have different motion characteristics even if they appear spatially close.
Data Source
AI summary
A LiDAR point cloud that includes two candidate clusters for merging is received. At a first phase, a distance between the two clusters is determined. If the distance is greater than a threshold, the candidate clusters are not merged. Otherwise, an additional point cloud is received for each cluster at different times. A motion characteristic is determined for each cluster. If the motion characteristic for each cluster is close (indicating that the objects are moving at the same speed), then the clusters are merged. Otherwise the clusters are not merged. The motion characteristic for a cluster can be determined by performing an alignment operation using the point cloud received for the cluster, and using the error associated with the alignment operation as the motion characteristic for the cluster. The decision to merge clusters is based on raw point cloud data, which can take place early in the tracking cycle.