Moving Robot Recognition Updates for Accurate Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot cleaner systems lack accurate obstacle detection due to the absence of machine learning capabilities, leading to inefficient navigation and potential collisions.
Innovation Solution
Integration of a machine learning framework between a moving robot and a server, where the robot generates field data from its environment and transmits it to the server for processing, allowing the server to generate update information for the robot's recognition algorithm to improve obstacle detection and navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning framework is integrated between robot and server, then obstacle detection accuracy is improved, but device complexity increases
Solution Approach 1:
A server is introduced as an intermediary between multiple robots to perform centralized machine learning operations. The server receives image data from robots, processes it using learning algorithms, and generates updated recognition algorithms that are transmitted back to robots. This mediator approach enables advanced obstacle detection without requiring complex hardware or processing power in each individual robot.
Solution Approach 2:
The patent replaces traditional mechanical obstacle detection methods (physical sensors, simple collision detection) with information-based machine learning algorithms. The system uses image processing and neural networks to detect and recognize obstacles, substituting physical/mechanical detection with computational intelligence that can identify obstacles before physical contact occurs.
2Productivity
If server-based learning is implemented, then navigation efficiency is improved, but loss of time in data transmission occurs
Solution Approach 1:
The server performs machine learning operations and generates updated recognition algorithms in advance, before the robots need them for navigation. By pre-processing image data and preparing improved algorithms during periods when robots are not actively navigating, the system minimizes the time robots spend waiting for updates, thereby maintaining high navigation efficiency while still benefiting from continuous learning.
Data Source
AI summary
A method of controlling a moving robot includes a travel control operation in which the moving robot generates field data by sensing information regarding an environment around the moving robot, and controls autonomous traveling based on a recognition result that is obtained by inputting the field data to a recognition algorithm for the moving robot; a server learning operation in which the moving robot transmits at least some of the field data to a server or other device and the server generates update information based on the field data transmitted by the moving robot; and an update operation in which the server transmits the update information to the moving robot, which uses the update information to update the recognition algorithm.