Weeding Robot Obstacle Recognition Without Boundary Wires
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing weeding robots require manual calibration using boundary wires, consuming significant manpower and material resources, and are limited by the shape constraints imposed by the burying of these wires, which complicates the weeding region's boundaries.
Innovation Solution
An obstacle recognition method and apparatus that determines a candidate obstacle region in a weeding region image using color, contour, and value information to identify obstacles without boundary wires, enhancing recognition efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If boundary wires are buried to calibrate boundaries of a weeding region, then the weeding robot can identify the working area, but a lot of manpower and material resources are consumed and costs are increased
Solution Approach 1:
The patent extracts the boundary calibration function from the physical boundary wires and transfers it to natural feature points in the environment (such as building corners, road intersections, and landscape features). This eliminates the need for boundary wires while maintaining the robot's ability to identify and calibrate its working area, thereby reducing manpower and material resources.
Solution Approach 2:
The patent introduces natural feature points as intermediaries between the robot and the environment. Instead of using artificial boundary wires, the robot uses these naturally occurring features (building corners, road intersections) as reference points for boundary calibration, achieving the same calibration function without the harmful boundary wire installation.
2Measurement precision
If boundary wires are buried to calibrate boundaries of a weeding region, then the working area can be defined, but a shape of the weeding region is limited to some extent
Solution Approach 1:
The patent removes the shape-constraining boundary wires and replaces them with flexible natural feature point-based calibration. This allows the robot to adapt to various irregular-shaped weeding regions by selecting appropriate natural feature points, eliminating the 90-degree corner limitation and enabling versatile shape accommodation.
Solution Approach 2:
The patent implements a dynamic boundary calibration system where the robot can flexibly select and adapt to different natural feature points based on the actual environment. This dynamic approach allows the weeding region shape to be adjusted and adapted to various layouts without being constrained by fixed boundary wire configurations.
3Productivity
If manual control is used for conventional weeding machines, then the weeding operation can be performed, but autonomous working functions are required for modern applications
Solution Approach 1:
The patent enables the weeding robot to perform autonomous boundary calibration by itself using natural feature points. The robot independently identifies feature points, calculates boundary information, and determines its working area without manual intervention, achieving self-service autonomous operation while maintaining effective weeding capability.
Data Source
AI summary
An obstacle recognition method includes the steps of: determining a candidate obstacle region in a candidate weeding region image according to color information of the candidate weeding region image; obtaining contour information of the candidate obstacle region and value information of the candidate weeding region image; and determining, according to the contour information and the value information, whether there is an obstacle in the candidate weeding region image. A related obstacle recognition apparatus, an electronic device, a computer-readable storage medium, and a weeding robot are also disclosed.