Robot Hand-Eye Calibration Using an On-Arm Calibrating Image
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hand-eye calibration methods for robotic manipulators are inefficient, requiring frequent recalibration and manual placement of calibration plates, which decreases working efficiency and is not applicable in all environments, especially with increased task variability and complex multi-hand robot systems.
Innovation Solution
A robot system with a calibrating image on the robotic manipulator, a camera, and a processor that moves the calibrating image towards the target object, records coordinate datasets, and executes a hand-eye calibration algorithm to obtain a calibrated mapping between camera and robotic manipulator coordinates, allowing for automated calibration without a calibration plate and in any environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional hand-eye calibration method with calibration plate is used, then calibration accuracy can be achieved, but working efficiency decreases due to frequent recalibration and manual operations
Solution Approach 1:
The calibration plate is extracted and replaced by a calibrating image disposed on the robotic manipulator itself. This eliminates the need for external calibration plates and manual placement, allowing the robot to perform self-calibration during task execution, thereby improving working efficiency while maintaining calibration accuracy.
Solution Approach 2:
The robotic manipulator performs hand-eye calibration on its own by using a calibrating image disposed on itself as the calibration target. The system automatically executes calibration procedures without manual intervention, enabling the robot to self-calibrate during task execution and eliminate the need for frequent external calibration operations.
2Adaptability or versatility
If manual placement of calibration plate is performed, then calibration can be executed, but the method is not applicable in all environments (e.g., dusted environments or when no space is available)
Solution Approach 1:
The external calibration plate is replaced by an on-board calibrating image disposed directly on the robotic manipulator. This extraction eliminates the dependency on external calibration objects, making the system adaptable to environments where calibration plates cannot be placed, such as dusted environments or spaces with limited room for calibration objects.
Solution Approach 2:
The calibrating image disposed on the robotic manipulator serves multiple functions: it acts as both a task object and a calibration target. This multi-functionality enables the system to perform calibration in any environment without requiring specific calibration conditions, thereby improving environmental adaptability while maintaining ease of operation.
3Reliability
If frequent hand-eye calibration is performed to handle task variability, then calibration accuracy is maintained, but time consumption increases
Solution Approach 1:
The hand-eye calibration is integrated into the task execution process, allowing calibration to occur continuously during normal robot operations. The robotic manipulator performs calibration by moving through its workspace while the camera captures images, transforming calibration from a discrete interruptive task into a continuous process that occurs alongside productive work, thereby maintaining calibration reliability without increasing time loss.
Solution Approach 2:
The system performs preliminary calibration actions by disposing the calibrating image on the robotic manipulator before task execution. This preliminary setup enables the robot to execute calibration procedures quickly during task variability events, as all calibration components are already in place and configured, reducing the time required for recalibration while maintaining calibration reliability.
4Measurement precision
If manual testing of various postures is performed, then comprehensive calibration data is collected, but the testing procedure takes a lot of time
Solution Approach 1:
The robotic manipulator automatically executes calibration procedures by moving through its workspace and collecting calibration data without manual intervention. The system self-manages the entire calibration process, including posture variation and data collection, eliminating the time-consuming manual testing of various postures while ensuring comprehensive calibration data is collected through automated workspace exploration.
Solution Approach 2:
Manual mechanical operations for posture testing and calibration plate placement are replaced by an automated vision-based system. The camera captures images during robot movement, and the processor automatically extracts calibration data, substituting manual mechanical testing with an automated optical-mechanical system that collects comprehensive calibration data more efficiently.
Data Source
AI summary
A robot for interacting with a target object includes a robotic manipulator, a calibrating image, a camera and a processor. The robotic manipulator corresponds to a robotic manipulator coordinate. The calibrating image is disposed on the robotic manipulator. The camera corresponds to a camera coordinate and for shooting the target object and generating a picture. The processor is configured to move the robotic manipulator such that the calibrating image moves towards the target object and enters the picture. The processor records robotic manipulator coordinate datasets and camera coordinate datasets of the calibrating image as the calibrating image moving towards the target object, and uses the robotic manipulator coordinate datasets and the camera coordinate datasets to execute a hand-eye calibrating algorithm to obtain a calibrated mapping between the camera coordinate and the robotic manipulator coordinate.