Vehicle LiDAR Outline Extraction with Dynamic Quadrant Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LiDAR systems face inaccuracies in object detection, leading to degraded reliability in autonomous driving, particularly due to the occurrence of unnecessarily continuous segments in object shape recognition.
Innovation Solution
An object outline extraction method that varies the reference for boundary quadrants based on the position of extracted objects, using a LiDAR sensor to detect and analyze point distributions, and resets the reference boundary to improve object shape recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed reference boundary is used for global quadrants, then the processing method is simple, but the object shape recognition accuracy deteriorates due to unnecessarily continuous segments
Solution Approach 1:
The patent applies the dynamics principle by making the reference boundary adjustable rather than fixed. The reference boundary is dynamically changed based on the distribution of point cloud data, allowing the system to adapt to different object shapes and positions. This resolves the contradiction by enabling accurate object shape recognition through dynamic adjustment while maintaining a relatively simple processing framework.
Solution Approach 2:
The patent changes the parameter of the reference boundary position based on point cloud distribution characteristics. By analyzing the distribution of detected points and adjusting the reference boundary accordingly, the system achieves more accurate object outline extraction. This parameter change approach allows the system to eliminate unnecessarily continuous segments while maintaining processing simplicity.
2Measurement precision
If the reference boundary is adjusted based on point distribution, then the object outline extraction accuracy is improved, but the processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-defining multiple possible reference boundary positions and their corresponding processing flows. The system analyzes point cloud distribution and selects the appropriate pre-prepared processing path, avoiding the need for complex real-time calculations. This resolves the time-accuracy contradiction by preparing multiple solutions in advance and selecting the most suitable one based on detected point distribution.
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 method enhances the accuracy of object shape recognition and reduces errors in map matching by preventing unnecessarily continuous segments, thereby improving the reliability of autonomous driving systems.
Implementation Method 1
Light detection and ranging (LiDAR) has been developed to construct and visualize geographic data for constructing three-dimensional (3D) geographic information system (GIS) information.
Data Source
AI summary
An object outline extracting method for a vehicle includes: detecting distributed positions of points in global quadrants generated based on a subject vehicle, as light detection and ranging (LiDAR) data is received; analyzing the distributed positions of the points and determining whether the points are distributed in a preset area based on a result value obtained from the analyzing; when it is determined that the points are distributed in the preset area, resetting a reference boundary of the global quadrants; and generating an object contour by connecting points found based on the reset reference boundary.