AI Cleaner Path Simplification for Obstacle-Dense Area Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robot cleaners often inefficiently clean areas with many obstacles, requiring prolonged cleaning times and repeated movements due to their inability to adapt cleaning paths based on the entire cleaning space and obstacles within a preset radius.
Innovation Solution
An artificial intelligence cleaner that modifies cleaning paths through path simplification by identifying candidate areas for simplification based on priority, using a SLAM map and machine learning algorithms to optimize cleaning efficiency and battery usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the robot cleaner considers only the environment within a preset radius from itself, then the device complexity is reduced, but the productivity deteriorates due to inefficient cleaning paths and repeated wandering motions
Solution Approach 1:
The cleaning space is segmented into multiple cleaning areas based on cleaning histories, allowing the robot to plan paths at different hierarchical levels - first selecting which area to clean next, then navigating within that area. This segmentation enables efficient global path planning while maintaining manageable local navigation complexity.
Solution Approach 2:
The robot performs preliminary analysis of cleaning histories to identify abnormal situations and pre-calculates optimized cleaning paths before execution. By anticipating problems like repeated wandering or excessive collisions in advance, the system can prepare alternative routes that improve productivity without increasing real-time navigation complexity.
2Reliability
If the robot cleaner performs cleaning while avoiding obstacles in the preset radius range, then the reliability is improved, but the loss of time increases due to repeated collisions and wandering motions in obstacle-dense areas
Solution Approach 1:
The system analyzes cleaning histories to identify areas with abnormal situations such as repeated collisions or excessive cleaning time in advance. Before executing the cleaning path, the processor pre-adjusts the path to avoid these problematic areas, eliminating time losses from repeated collisions while maintaining reliable obstacle avoidance through historical pattern recognition.
Solution Approach 2:
The robot uses cleaning history data as feedback to continuously improve path planning. By monitoring which areas require repeated cleaning attempts or excessive time, the system learns from past performance and adjusts future paths to avoid recurring time losses, thereby improving both reliability and efficiency over time.
3Ease of operation
If the robot cleaner cleans areas with many obstacles using preset radius-based navigation, then the ease of operation is maintained, but the productivity deteriorates due to restricted movement and locked navigation
Solution Approach 1:
The cleaning space is divided into multiple cleaning areas based on historical data, allowing the robot to handle obstacle-dense regions as separate manageable units. This segmentation enables the system to maintain simple operation within each area while improving overall productivity by optimizing the sequence and path between areas, avoiding restricted movement through global path planning.
Solution Approach 2:
The cleaning path is dynamically adjusted based on cleaning histories and identified abnormal situations. The processor modifies paths in real-time to avoid areas where movement is restricted or locked due to obstacles, maintaining ease of operation through automatic adaptation while significantly improving cleaning speed by preventing unproductive wandering motions.
Data Source
AI summary
An embodiment of the present invention provides an artificial intelligence cleaner comprising: a memory for storing a simultaneous localization and mapping (SLAM) map of a cleaning space; a travel driving part for driving the artificial intelligence cleaner; and a processor for collecting a plurality of cleaning records for the cleaning space, dividing the cleaning space into a plurality of cleaning areas by using the SLAM map and the plurality of collected cleaning records, determining a cleaning path of the artificial intelligence cleaner in consideration of the divided cleaning areas, and controlling the travel driving part according to the determined cleaning path, wherein when an abnormal situation occurs during cleaning on the basis of the determined cleaning path, the processor modifies the cleaning path by applying path simplification to a preconfigured area of the remaining cleaning area.