Mobile robot and method of controlling the same
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile robot technologies face challenges in accurately recognizing location and creating maps in environments with varying lighting conditions and object changes, leading to reduced accuracy in location estimation and map creation.
Innovation Solution
A mobile robot system that combines vision-based location recognition using a camera with LiDAR-based location recognition, utilizing a controller to integrate data from both sensors to create a robust map and accurately estimate location, even in changing environments, by employing odometry information and iterative closest point (ICP) matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vision-based location recognition using a camera is used, then location recognition can be performed, but accuracy varies due to environmental changes such as lighting conditions and object location changes
Solution Approach 1:
The patent combines vision-based location recognition (using camera images and feature points) with LiDAR-based location recognition (using laser ranging and point cloud matching) into a unified system. The controller integrates both recognition results to determine the mobile robot's location, thereby compensating for the weaknesses of each individual method and achieving accurate and robust location recognition under various environmental conditions.
2Ease of operation
If feature point matching is used for location recognition, then location can be estimated, but success rate decreases in environments with lighting changes or object movements
Solution Approach 1:
The patent introduces LiDAR-based location recognition as an intermediary method to support and verify vision-based location recognition. When the camera-based feature point matching fails or produces unreliable results (indicated by low success rate), the LiDAR system provides alternative location estimation through point cloud matching, ensuring continuous and reliable operation.
3Device complexity
If only camera-based vision recognition is used, then the system remains simple, but location accuracy deteriorates in changing environmental conditions
Solution Approach 1:
The patent employs a composite sensing system that integrates two different types of sensors (camera and LiDAR) with complementary characteristics. The camera provides rich visual information for feature recognition, while the LiDAR provides accurate depth and geometric information. This composite approach creates a more robust location recognition system that maintains high accuracy across diverse environmental conditions.
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 system achieves reliable and accurate location recognition and map creation, enhancing the success rate of location estimation and enabling efficient navigation and cleaning operations across various environmental conditions.
Implementation Method 1
a LiDAR sensor for acquiring distance information on the basis of a reflected beam of light
Data Source
AI summary
A mobile robot includes a traveling unit configured to move a main body, a LiDAR sensor configured to acquire geometry information, a camera sensor configured to acquire an image of the outside of the main body, and a controller. The controller generates odometry information based on the geometry information acquired by the LiDAR sensor. The controller determines a current location of the mobile robot by performing feature matching between images acquired by the camera sensor based on the odometry information. The controller combines the information obtained by the camera sensor and the LiDAR sensor to accurately determine the current location of the mobile robot.