Autonomous Robot Map Management for Repositioning and Fast Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous mobile robots face inefficiencies and delays due to the need for repeated map creation and time-consuming global self-localization in changing environments, especially in multi-floor households where they must handle multiple maps and adapt to moving objects and furniture.
Innovation Solution
The method involves storing and managing maps for different robot application areas, allowing the robot to navigate using stored maps and perform tasks efficiently, while creating new maps during tasks that require global self-localization, and using a working copy of the map to adapt to environmental changes without altering the permanent map.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the robot creates a new map with each work process (temporary map), then the robot can avoid storing multiple permanent maps, but the robot must repeatedly explore the environment to build maps, reducing efficiency
Solution Approach 1:
The patent divides the map management system into two segments: a permanent map storage system that stores environment information across multiple work processes, and a temporary working map that is created during each specific task. This segmentation allows the robot to reuse permanent map data while maintaining flexibility for current tasks, eliminating the need to completely rebuild maps each time.
Solution Approach 2:
The patent implements preliminary action by storing permanent map information about the environment before work processes begin. This pre-stored information includes environment-specific data that can be reused across multiple tasks, so the robot does not need to repeatedly explore and map the entire environment during each work process.
2Productivity
If the robot uses permanently saved maps, then the robot can navigate more efficiently without repeated exploration, but the robot must perform time-consuming global self-localization to determine which map it is in
Solution Approach 1:
The patent applies local quality by creating a working map that is locally adapted to the current task and environment state, rather than using a single global permanent map. The working map contains only the relevant information needed for the current work process, allowing faster localization and navigation without processing entire permanent maps.
Solution Approach 2:
The patent extracts the necessary navigation information from the permanent map to create a simplified working map for the current task. This extraction process removes unnecessary details and focuses only on relevant environment-specific information, reducing the time required for self-localization while maintaining navigation accuracy.
3Adaptability or versatility
If the robot uses permanently saved maps, then the robot can reuse environment information across tasks, but the robot must handle multiple maps and determine which map applies to the current location
Solution Approach 1:
The patent performs preliminary action by pre-processing permanent map data to create working maps that are ready for immediate use in specific tasks. This preparation work is done in advance, organizing environment information in a task-specific format that simplifies the robot's decision-making process during actual work.
Solution Approach 2:
The patent uses copying by creating a working map that is a selective copy of the permanent map, tailored to the current task requirements. This copy contains only the relevant portion of the permanent map data, making it easier to handle and process while preserving the ability to reuse environment information from the original permanent map.
4Measurement precision
If the robot performs global self-localization by checking the new map against stored maps, then the robot can accurately determine its location, but the comparison process delays task execution
Solution Approach 1:
The patent applies local quality by focusing the self-localization process on the working map which contains only locally relevant information for the current task. Instead of comparing the new map against entire permanent maps, the robot performs comparison on the condensed, task-specific working map data, maintaining accuracy while reducing computation time.
Solution Approach 2:
The patent implements partial action by performing self-localization comparison only on the necessary portions of map data that are relevant to the current task and location. Rather than exhaustively comparing all permanent map data, the robot performs a targeted partial comparison that achieves sufficient accuracy without excessive time consumption.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
Described is, inter alia, a method for controlling an autonomous mobile robot. According to one exemplary embodiment, the method comprises the storage and management of at least one map associated with an area of use for the robot and the navigation of the robot through said area of use for the robot, wherein the robot continuously determines its position on the map. The method further comprises the detection of a repositioning procedure, during which the robot carries out a movement that the robot itself cannot control. During this repositioning procedure, the robot detects information about its position and/or its state of motion with the aid of sensors and, based on the detected information, determines an estimated value for its position.