Mobile robot and control method therefor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mobile robots lack the ability to accurately determine the attribute of objects, leading to unreliable object recognition and avoidance, which affects their driving and cleaning performance.
Innovation Solution
A mobile robot equipped with an image acquisition unit, sensor unit, and a controller featuring an object recognition module that uses machine learning to recognize object attributes by acquiring images and sensing data, allowing for accurate object classification and adjustment of travel patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional sensors (infrared or ultrasonic) are used to detect objects, then the mobile robot can sense object presence and distance, but it cannot determine the attribute of the object
Solution Approach 1:
The patent combines multiple sensing approaches by integrating image acquisition (camera) with traditional sensor data (infrared/ultrasonic). The controller merges image recognition results with sensor detection data to comprehensively determine both object presence and attributes, resolving the limitation of single-sensor systems that can detect but not classify objects.
Solution Approach 2:
The patent replaces or supplements traditional mechanical sensing methods with optical image acquisition and machine learning-based recognition. Instead of relying solely on infrared or ultrasonic sensors that provide limited data, the system uses image cameras combined with neural network algorithms to extract detailed object attributes such as type, size, and characteristics.
2Productivity
If the mobile robot moves forward when an object is detected with climbable height, then it can efficiently clean the area, but it may be caught or restricted by objects like electric wires
Solution Approach 1:
The system implements feedback control by continuously analyzing image data and sensor information to determine object attributes before deciding on movement actions. The controller uses recognized object attributes (such as identifying electric wires versus climbable thresholds) to feedback-adjust the traveling pattern, ensuring both cleaning efficiency and safety by avoiding objects that could restrict movement.
Solution Approach 2:
The patent changes the decision-making parameters from simple height-based classification to comprehensive attribute recognition. By using machine learning to identify object types (electric wire, threshold, fan base, etc.), the system adjusts its traveling parameters dynamically - moving forward for safe objects and avoiding or detouring for hazardous objects, thus resolving the contradiction between productivity and reliability.
3Device complexity
If the mobile robot uses simple height-based object classification, then the control logic is simple, but it cannot distinguish between different object types with similar heights
Solution Approach 1:
The patent introduces an image recognition module as an intermediary between sensor detection and movement control. This intermediary layer processes visual information to classify object types, providing detailed attribute information that bridges the gap between simple detection and complex decision-making, enabling accurate differentiation of objects with similar heights while maintaining systematic control architecture.
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
The mobile robot achieves highly reliable object recognition and avoidance, enhancing user convenience, operational efficiency, and cleaning efficiency by accurately determining object attributes and adjusting its movement accordingly.
Implementation Method 1
The infrared sensor determines presence of an object and a distance to the object based on an amount of light reflected by the object
Implementation Method 2
the ultrasonic sensor emits an ultrasound wave having a predetermined period and, in response to ultrasonic waves reflected by an object, determines a distance to the object based on a difference between the time to emit the ultrasonic waves and the time for the ultrasonic waves to return to the object
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A mobile robot in one general aspect of the present invention includes: a travel drive unit configured to move a main body; an image acquisition unit configured to acquire an image of surroundings of the main body; a sensor unit having one or more sensors configured to sense an object during the movement of the main body; a storage configured to, when the sensor unit senses an object, store position information of the sensed object and position information of the mobile robot, register an area having a predetermined size around a position of the sensed object as an object area in a map, and store the image acquired by the image acquisition unit in the object area; and a controller having an object recognition module configured to recognize the object sequentially with respect to images acquired by the image acquisition unit in the object area, and determine a final attribute of the object based on a plurality of recognition results of the sequential recognition, and accordingly, it is possible to accurately recognize an attribute of an object and register and manage an object area.