Robot cleaner and control method thereof
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robot cleaners face challenges in accurately recognizing their position on a map and returning to a charging base after being moved by external factors, leading to potential battery depletion due to inefficient search methods for infrared signals, highlighting the need for a global localization method.
Innovation Solution
A control method for robot cleaners that involves acquiring images of surroundings, estimating room-specific feature distributions, and comparing these with stored reference distributions to determine the current position, using image acquisition units and feature distribution learning modules to create and match label-specific score distributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the robot cleaner uses infrared signal sensing to search for the charging base, then the robot can locate the charging base, but the robot may wander around and deplete battery when its position is changed by external factors
Solution Approach 1:
The robot cleaner performs preliminary actions by creating and storing a map of the cleaning area including charging base position information before needing to return. When the robot's position is changed, it can quickly re-localize by comparing current sensor data with the pre-stored map rather than wandering to search for the charging base, thus preventing unnecessary battery consumption.
Solution Approach 2:
The system implements feedback by continuously monitoring the robot's position relative to the stored map and adjusting its navigation accordingly. When position change is detected, the robot uses sensor feedback (cameras, sensors) to re-localize itself on the map and recalculate the path to the charging base, ensuring reliable return while minimizing energy waste.
2Measurement precision
If the robot cleaner creates a map of the cleaning area, then the robot can recognize positions, but the robot cannot recognize position when forcibly moved to another room
Solution Approach 1:
The map creation system is designed to be universal across multiple rooms and cleaning areas. The robot creates a comprehensive map that includes multiple rooms and the charging base position, allowing it to recognize its position regardless of which room it is in. This multi-functional map structure enables the robot to adapt to position changes and still accurately locate itself and navigate to the charging base.
Solution Approach 2:
The robot performs preliminary exploration to create a complete map of all cleaning areas including multiple rooms before operational use. This pre-established comprehensive map serves as a reference that allows the robot to quickly determine its position even after being moved to another room, eliminating the need to recreate the map and enabling immediate re-localization.
3Reliability
If the robot cleaner searches for charging base without knowing position, then the robot can attempt to locate charging base, but the robot runs out of battery while wandering
Solution Approach 1:
The robot performs preliminary actions by creating and storing a detailed map of the cleaning area including the charging base position before any cleaning or charging operations. This pre-established map eliminates the need for time-consuming searches, as the robot can directly calculate its position and navigate to the charging base using the stored map information, significantly reducing search time and preventing battery depletion.
Solution Approach 2:
The system replaces the mechanical wandering search method with an information-based navigation approach. Instead of physically moving around to search for the charging base using infrared sensors, the robot uses pre-stored map data and current position information to calculate and directly navigate to the charging base, substituting computational navigation for mechanical exploration and dramatically reducing time loss.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Disclosed herein is a control method for a robot cleaner including acquiring a plurality of images of surroundings during travel in a cleaning area, estimating at least one room-specific feature distribution according to a rule defined for each of at least one room, based on the images acquired in the acquiring of the plurality of images, acquiring an image of the surroundings at a current position, obtaining a comparison reference group including a plurality of feature distributions by applying the rule for each of the at least one room to the image acquired in the acquiring of the image at the current position, and comparing the obtained comparison reference group with the estimated at least one room-specific feature distribution and determining a room having the robot cleaner currently positioned therein.