LIDAR Object Recognition Using Representative Point Contour Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object recognition systems for autonomous vehicles face challenges in accurately identifying contour points from a cloud of points obtained through LIDAR, while efficiently reducing computational resources and maintaining accuracy.
Innovation Solution
An object recognition apparatus and method that extracts representative points from LIDAR data, determines contour points based on distance and angle thresholds, and controls vehicles accordingly, thereby enhancing accuracy and reducing computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all points from LIDAR data are processed to identify contour points, then the accuracy of object shape identification is improved, but the computational load and processing time increase significantly
Solution Approach 1:
The patent segments the dense point cloud data into a simplified set of representative points that capture the essential contour information. Instead of processing all LIDAR points, the system selects key representative points that define the object's shape boundaries, thereby reducing computational load while maintaining identification accuracy.
Solution Approach 2:
The patent extracts only the necessary contour-defining points from the complete LIDAR point cloud. By taking out and focusing solely on representative points that lie on or near the object's contour, the system eliminates redundant internal points while preserving the accuracy needed for shape identification.
2Productivity
If a simplified method is used to identify contour points, then the computational load is reduced, but the accuracy of contour point identification deteriorates
Solution Approach 1:
The patent applies local quality by using different selection criteria for different regions of the point cloud. Representative points are selected based on their local geometric properties and their contribution to defining the contour, ensuring that each selected point serves a specific purpose in accurately representing the object's shape at its local location.
Solution Approach 2:
The patent changes parameters such as the density and distribution of selected representative points based on the local complexity of the object's contour. In regions with high curvature or complex geometry, more representative points are selected, while in simpler regions, fewer points suffice, thereby maintaining accuracy while optimizing computational efficiency.
3Reliability
If more representative points are selected as contour points, then the accuracy of object recognition is improved, but the amount of data to be processed increases
Solution Approach 1:
The patent applies partial action by selecting only the minimum necessary number of representative points required to reliably define the object's contour and shape. Rather than using all available points or an excessive number, the system identifies and processes only those points that are critical for accurate object recognition, thereby maintaining reliability while minimizing data volume.
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
The proposed solution improves the accuracy of contour point identification, reduces computational requirements, and enhances the reliability of object recognition for autonomous vehicles, leading to better decision-making and safety in navigation.
Implementation Method 1
A distance from a LIDAR to an object can be measured based on an interval between the time when laser is transmitted by the LIDAR and the time when the laser reflected off the object is received
Data Source
AI summary
The present disclosure relates to an object recognition apparatus and method. The object recognition apparatus includes a sensor (e.g., LIDAR) and a processor. The processor may extract a plurality of representative points, determine a first reference contour point, a first representative point, a second representative point, and a third representative point among the plurality of representative points, determine a first distance between the first reference contour point and the second representative point; determine a first angle between a first ray extending from the first reference contour point to the first representative point and a second ray extending from the second representative point to the third representative point, determine a fourth representative point, and determine, as a contour point, one of the first representative point, the second representative point, or the fourth representative point as the contour point.