A High-Precision Hand-Eye Coordination Localization Method under Convex Relaxation Optimization

Area of Science:

• Robotics
• Electrical Engineering
• Computer Vision

Background:

• Traditional switching operations pose risks due to on-site work and high-voltage arc discharges.
• Existing hand-eye calibration methods struggle with coupled rotation and translation solutions, impacting precision.

Purpose of the Study:

• To develop a robust and accurate hand-eye calibration algorithm for robot-assisted electrical switching.
• To improve the safety and precision of remote switching operations.

Main Methods:

• Utilized dual quaternions to simplify the hand-eye calibration model.
• Applied convex relaxation and linear matrix inequality convex optimization for accurate rotation matrix determination.
• Developed a new objective function to address coupling issues between rotation and translation vectors.

Main Results:

• Achieved a positioning precision of approximately 2.9 mm.
• Demonstrated superior calibration accuracy, noise robustness, and stability compared to existing algorithms.
• Validated performance through experiments with added Gaussian noise.

Conclusions:

• The proposed dual quaternion-based convex relaxation algorithm significantly enhances hand-eye calibration for robotic switching.
• The algorithm meets the stringent accuracy requirements for electrical switching operations, improving safety and efficiency.