Robot Vision Servo Gripping in Uncalibrated Agricultural Scenes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic technologies require precise calibration and are not adaptable to unstructured and complex agricultural scenes, limiting their applicability and environmental adaptability.
Innovation Solution
A method and device for collaborative servo control of motion vision in uncalibrated agricultural scenes using a robot arm with a mechanical gripper, image sensor, and control module, which constructs scene space feature vectors and inverse reinforcement reward policy networks for guided policy search, allowing the robot to grip targets without precise spatial calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If precise calibration is performed to acquire spatial coordinates of target endpoint, then manufacturing precision is improved, but device complexity and calibration requirements increase
Solution Approach 1:
The patent replaces traditional mechanical calibration systems with a vision-based approach. The robot uses image processing and visual feedback to locate and grip targets without requiring precise pre-calibration of spatial coordinates. The control module processes visual information to determine target position and guides the robot arm accordingly, substituting complex mechanical calibration with simpler optical sensing and computational methods.
2Manufacturing precision
If traditional trajectory programming is used with pre-calculated ray intersections, then manufacturing precision is improved, but adaptability to different scenes deteriorates
Solution Approach 1:
The patent implements dynamic trajectory adjustment based on real-time visual feedback. Instead of relying on pre-calculated fixed trajectories, the system continuously monitors target position through image sensing and dynamically adjusts the robot arm's movement path. This allows the system to adapt to different scenes and target positions while maintaining gripping accuracy, combining the benefits of precise control with environmental adaptability.
Solution Approach 2:
The system employs visual feedback loops where the image sensor continuously captures target position, the control module processes this information to determine deviations from the planned trajectory, and the robot arm adjusts its motion accordingly. This feedback mechanism enables the system to maintain precision while adapting to unstructured agricultural scenes without requiring pre-calibration.
3Measurement precision
If visual servo control with multiple coordinate systems is implemented, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent extracts and eliminates the complex coordinate system transformation step from the control process. Instead of requiring multiple coordinate system conversions (earth coordinate system, UAV coordinate system, camera coordinate system), the system directly uses image coordinates from the sensor to guide the robot arm. The control module processes visual data in a simplified manner, removing unnecessary intermediate transformation steps while maintaining measurement precision through direct visual-feedback-based positioning.
Data Source
AI summary
A device and method for collaborative servo control of motion vision of a robot in an uncalibrated agricultural scene is provided. The device includes a robot arm, a to-be-gripped target object, an image sensor and a control module. An end of a robot arm is provided with a mechanical gripper, and a to-be-gripped target object is within a grip range of the robot arm. A control module drives the mechanical gripper to grip the to-be-gripped target object, and controls an image sensor to perform image sampling on a process of gripping the to-be-gripped target object by the robot arm. The image sensor sends sampled image data to the control module. The device does not need to perform precise spatial calibration on the to-be-gripped target object and the related environment in the scene. The robot arm is guided to complete the gripping task according to trained networks.