Point Cloud Edge Extraction for Accurate Robot Along-Wall Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LiDAR-based systems for robot navigation have low accuracy in three-dimensional perception, leading to inaccurate along-edge semantic information and increased risk of collisions with obstacles not at the LiDAR's installation height, and require costly manual deployment for precise path planning.
Innovation Solution
A method involving a server or terminal system that processes point cloud data to extract wall surface points, projects them to a ground surface, performs linear fitting, and combines vectors to generate accurate along-edge semantic information, enhancing obstacle avoidance capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If LiDAR is used for edge detection, then the robot can navigate along edges, but the three-dimensional perception ability is insufficient leading to low accuracy
Solution Approach 1:
The patent combines LiDAR point cloud data with camera images to create a fused data structure that leverages the complementary strengths of both sensors. The LiDAR provides geometric information while the camera provides semantic labeling, achieving both edge detection capability and three-dimensional perception accuracy.
Solution Approach 2:
The patent introduces a server system as an intermediary that processes point cloud data from LiDAR and combines it with image data from cameras. This server-based processing enables accurate along-edge semantic information generation by integrating multiple data sources that would be insufficient when used separately.
2Device complexity
If LiDAR point cloud data is used directly, then the system is simple, but obstacles at different heights cannot be perceived leading to collisions
Solution Approach 1:
The patent merges LiDAR point cloud data with camera image data to create a comprehensive obstacle detection system. The camera captures obstacles at various heights while the LiDAR provides spatial positioning, enabling reliable obstacle avoidance without requiring complex individual sensor systems.
Solution Approach 2:
The server system performs multiple functions including point cloud processing, image data integration, obstacle detection, and path planning. This multi-functional approach improves reliability by consolidating processing capabilities while maintaining system manageability.
3Measurement precision
If manual deployment is used for precise path planning, then accuracy is achieved, but deployment costs increase
Solution Approach 1:
The server system automatically processes point cloud data and generates accurate along-edge semantic information without requiring manual deployment or configuration. The system self-calibrates and adapts to different environments, achieving precise path planning accuracy while eliminating costly manual setup procedures.
Solution Approach 2:
The patent replaces manual deployment methods with automated computer-based processing. The server system uses algorithms to process sensor data and generate navigation paths, substituting mechanical/manual operations with digital processing that is both more accurate and cost-effective.
Data Source
AI summary
A point cloud data processing method, an electronic device, and a computer-readable storage medium are provided. The method includes: obtaining a plurality of obstacle point clouds, and extracting wall surface point cloud(s) from the plurality of obstacle point clouds; obtaining a first point cloud vector by projecting the obstacle point clouds in the plurality of obstacle point clouds other than the wall surface point cloud(s) to a ground surface; projecting the wall surface point cloud(s) to the ground surface, and obtaining a straight line by performing a linear fitting on the projected wall surface point cloud(s); obtaining a second point cloud vector by screening the projected wall surface point cloud(s) based on the straight line; and combining the first point cloud vector and the second point cloud vector into along-edge semantic information. Through present disclosure, more accurate along-edge semantic information can be obtained.