Mobile Robot Position Recognition Using Stereo Camera Direct Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile robots face challenges in accurately determining their position due to errors in dead reckoning navigation, especially when using monocular cameras, and the high cost of laser scanners, which limits their mobility and service capabilities.
Innovation Solution
A position recognition method using direct tracking, which projects seeds onto stereo images to calculate intensity differences, updates camera status parameters, and estimates position based on odometry information from visual or inertial data, reducing positional errors and eliminating the need for expensive sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If monocular camera is used for position recognition, then device cost is reduced, but measurement precision deteriorates due to inability to determine distance to feature points
Solution Approach 1:
The patent transitions from monocular (2D) camera to stereo camera (adding depth dimension) to enable accurate distance measurement. By introducing a second camera viewpoint, the system calculates real-world distances through triangulation, resolving the precision limitation of single-camera systems while maintaining cost-effectiveness compared to laser scanners.
Solution Approach 2:
The patent uses a stereo camera system that creates a visual copy of the environment from multiple viewpoints. By projecting seeds from key frames onto current stereo images and comparing intensity differences, the system reconstructs three-dimensional spatial information without requiring expensive dedicated distance sensors.
2Measurement precision
If laser scanner is used for position recognition, then measurement precision is improved, but device cost increases significantly
Solution Approach 1:
The patent replaces expensive laser scanners with relatively inexpensive stereo camera systems. While cameras have limitations in certain conditions, they provide sufficient precision for mobile robot navigation at a fraction of the cost of laser scanners, making the system economically viable for widespread deployment.
Solution Approach 2:
The patent substitutes mechanical/optical laser scanning systems with visual-based stereo imaging and computational processing. By using computer vision algorithms to process stereo images and calculate positions, the system achieves comparable precision to laser scanners while eliminating the need for expensive specialized hardware.
3Device complexity
If dead reckoning navigation is used, then device complexity is reduced, but reliability deteriorates due to accumulated errors
Solution Approach 1:
The patent implements feedback by continuously comparing the robot's predicted position (from dead reckoning) with actual visual observations from stereo cameras. The system projects seeds from key frames onto current images, calculates intensity differences, and updates position estimates accordingly, correcting accumulated errors in real-time while maintaining system simplicity.
Solution Approach 2:
The patent merges dead reckoning navigation with visual odometry from stereo cameras. By combining the simplicity of dead reckoning with the accuracy of visual feedback, the system achieves reliable position recognition without requiring complex individual systems, as the two methods complement each other's strengths and weaknesses.
Data Source
AI summary
Disclosed are an apparatus of recognizing a position of a mobile robot using direct tracking and a method thereof. An apparatus of recognizing a position of a mobile robot includes two wide angle cameras which obtain one pair of stereo images on a region of interest including a vertical direction and a horizontal direction in accordance with movement of a mobile robot; and a position recognizing unit which extracts an edge from the received stereo image to project a seed in a key frame selected based on the extracted edge onto the stereo images, updates a status parameter indicating rotation and movement of a camera using an intensity error in the calculated projected seed as the projection result, and estimates a position of the mobile robot using the update result.