Robot Boundary-Walking Area Cleaning for Mapless Real-Time Coverage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot cleaning methods face inefficiencies, such as excessive navigation paths and slow cleaning due to mismatched terrain outlines, and require prior maps for effective area coverage, which can lead to redundant map construction and navigation path calculation.
Innovation Solution
An area cleaning planning method where a robot walks along the boundary of predefined cleaning areas, using real-time laser maps to divide and clean room subareas matching the actual environment, without pre-storing a global map, ensuring efficient navigation and coverage by automatically dividing areas as it cleans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the robot uses rectangular frame area cleaning mode without prior map, then no prior map is needed, but the difference between framed area outline and actual terrain is large causing excessive navigation paths and slow cleaning
Solution Approach 1:
The patent applies dynamics by transitioning from static rectangular frame areas to dynamic matched areas that adapt to actual terrain boundaries. The robot dynamically identifies room boundaries during cleaning and adjusts cleaning areas to match actual room shapes, allowing the system to maintain both mapless operation and high cleaning efficiency.
Solution Approach 2:
The patent changes the parameter of area shape from fixed rectangular frames to variable matched areas that conform to actual room boundaries. This parameter change allows the cleaning path to adapt to terrain features, reducing unnecessary navigation while maintaining the advantage of not requiring prior maps.
2Productivity
If the robot uses prior map based room division mode, then cleaning speed is accelerated and navigation paths are reduced, but prior map is required which increases system complexity
Solution Approach 1:
The patent applies preliminary action by performing boundary identification and area matching during the cleaning process itself rather than requiring pre-processing. The robot proactively detects walls and obstacles during navigation and uses this information to dynamically define cleaning areas, achieving map-based efficiency without the complexity of prior map creation.
3Loss of information
If the robot walks along boundary for map construction, then complete area coverage is achieved, but prolonged boundary walking reduces cleaning efficiency
Solution Approach 1:
The patent merges boundary detection and cleaning operations into a single integrated process. Instead of separating map construction from cleaning, the robot performs both simultaneously by identifying room boundaries during cleaning navigation, eliminating redundant boundary walking and achieving complete coverage without time loss.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed are an area cleaning planning method for robot walking along the boundary, a chip and a robot. The area cleaning planning method includes: on a laser map which is scanned and constructed by a robot in real time, the robot is controlled to walk along the boundary in a predefined cleaning area framed at the current planning starting point position, so that the robot does not cross out the predefined cleaning area in the process of walking along the boundary; meanwhile, according to the division condition of the room cleaning subareas that conform to the preset wall environment condition in the predefined cleaning area , the robot is controlled to walk along the boundary in a matched area, when the robot walks along the boundary in the matched area and returns to the planning starting point position, the robot is controlled to perform planned cleaning in the matched area; and when the robot finishes planned cleaning in the matched area, the robot is controlled to move to the next planning starting point position of a non-cleaned area, and the steps are repeated until no non-cleaned areas are detected.