Mobile robot and control method therefor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D sensor-based robots struggle to accurately detect obstacles with thin shapes, leading to inefficient and inappropriate driving and cleaning operations.
Innovation Solution
A mobile robot accumulates sensing results from a 3D camera sensor over a predetermined time period to generate face information, using a recognition model to detect obstacles, and controls driving based on this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a 3D camera sensor uses line light for obstacle detection, then the detection speed is improved, but the detection accuracy deteriorates for thin-shaped obstacles
Solution Approach 1:
The system performs preliminary scanning of the environment using line light to identify potential obstacle regions before conducting detailed detection. This allows the robot to pre-position itself or pre-activate additional sensors for areas where thin-shaped obstacles are suspected, improving both speed and accuracy by preparing detection resources in advance.
Solution Approach 2:
The system transitions from two-dimensional line light detection to three-dimensional detection by incorporating depth information and spatial mapping. This dimensional enhancement allows the robot to detect thin-shaped obstacles by analyzing their spatial context and relationships with surrounding objects, overcoming the limitation of line light while maintaining efficient detection.
2Area of stationary object
If a 3D camera sensor is used for obstacle detection, then the detection range is improved, but the detection accuracy for thin-shaped obstacles deteriorates
Solution Approach 1:
The detection field is segmented into multiple zones with different detection strategies. High-priority zones where thin-shaped obstacles are likely to exist use enhanced detection algorithms and multiple sensor types, while lower-priority zones use standard line light detection. This segmentation maintains wide coverage while improving accuracy in critical areas.
Solution Approach 2:
The system introduces intermediary processing steps between wide-area scanning and final obstacle identification. These intermediaries include preliminary classification of detected objects, spatial relationship analysis, and contextual verification that help distinguish thin-shaped obstacles from background elements, improving accuracy without reducing detection range.
3Loss of time
If obstacle detection relies on single sensor reading, then the response time is improved, but the detection reliability deteriorates
Solution Approach 1:
The system implements periodic detection cycles with increasing frequency based on operational context. During normal operation, detection occurs at standard intervals. When the robot approaches areas with historical obstacle data or detects potential obstacle signatures, the detection frequency increases automatically. This periodic action maintains fast response times while improving reliability through repeated verification.
Solution Approach 2:
The system incorporates feedback mechanisms where detection results from previous cycles inform subsequent detection strategies. If thin-shaped obstacles are detected or suspected, the system adjusts its detection parameters, activates additional sensors, or modifies its movement pattern to allow for better observation. This feedback loop improves reliability without significantly increasing response time by making detection adaptive rather than uniformly frequent.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the accuracy and reliability of obstacle detection, particularly for thin-shaped obstacles, improving driving stability and usability of the 3D camera sensor.
Implementation Method 1
a 3D camera sensor may be provided to detect an obstacle
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The present specification relates to a mobile robot and a control method therefor, in which the mobile robot generates virtual floor surface information about a travelling environment by accumulating, for a certain time, the results of sensing via a camera sensor to improve the accuracy of obstacle detection using the camera sensor, detects whether or not there is an obstacle in the travelling environment, on the basis of the floor surface information, and controls travel according to the result of the detection.