Mapping negative space for autonomous mobile robot
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous mobile cleaning robots face challenges in accurately mapping environments, particularly when large non-traversable items like beds are present, leading to unrecognizable visual displays of maps, which complicates navigation and cleaning operations.
Innovation Solution
The use of additional sensor data to modify the boundary of the environment, allowing for a more accurate representation of non-traversable spaces and the generation of three-dimensional maps, which helps users recognize the layout and improves the robot's cleaning efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the robot maps the environment using traditional boundary detection methods, then the mapping coverage is complete, but the visual display of the map becomes unrecognizable when large items like beds are present
Solution Approach 1:
The patent segments the mapping process into two distinct components: traversable space boundary (based on robot navigation capability) and non-traversable space boundary (based on object detection). This segmentation allows the map to distinguish between areas the robot can clean and areas occupied by objects, resolving the contradiction between complete mapping coverage and map recognizability by presenting a simplified, recognizable layout to users while maintaining accurate navigation data internally
Solution Approach 2:
The patent introduces an intermediary processing layer that transforms raw sensor data and navigation boundaries into a user-friendly map representation. This intermediary process filters out non-traversable spaces (objects) from the visual display while preserving them in the navigation data, acting as a mediator between the robot's precise measurement capabilities and the user's need for recognizable maps
2Loss of information
If the robot uses sensor data to detect all objects in the environment, then the mapping detail increases, but the complexity of processing and displaying the data increases
Solution Approach 1:
The patent extracts only the essential information needed for navigation and cleaning operations from the full sensor data set. By separating traversable from non-traversable spaces and focusing on boundary detection rather than complete object characterization, the system retains critical environmental details while eliminating unnecessary data processing complexity
Solution Approach 2:
The patent applies partial action by detecting and processing only the portions of the environment that are relevant to cleaning operations (traversable spaces and their boundaries). Rather than fully processing and displaying all detected objects and details, the system focuses on the essential boundaries needed for navigation, reducing data processing complexity while maintaining sufficient environmental awareness
3Measurement precision
If the robot creates a detailed map including all non-traversable spaces, then the map accuracy improves, but the setup time and user configuration time increase
Solution Approach 1:
The patent enables the robot to automatically distinguish and categorize spaces as traversable or non-traversable through its own sensor data and navigation experience, without requiring user intervention to manually configure each object. The robot serves itself by autonomously creating the boundary classification, maintaining accurate environmental representation while minimizing setup time and user configuration requirements
Data Source
AI summary
A non-transitory machine-readable medium, including instructions, which when executed, cause processing circuitry to perform operations to receive sensor data from a mobile cleaning robot based on interactions between the mobile cleaning robot and an environment. The instructions can further cause the processing circuitry to perform operations to generate a boundary of traversable space by the mobile cleaning robot within the environment using the sensor data, the boundary at least partially defining non-traversable space of the environment, and the non-traversable space including a region beyond the boundary. The instructions can further cause the processing circuitry to perform operations to generate a modified boundary of the environment using the non-traversable space and the sensor data.