Methods for Simultaneous Robot-World-Hand-Eye Calibration: A Comparative Study.

Area of Science:

• Robotics
• Computer Vision
• Calibration Techniques

Background:

• Accurate robot calibration is crucial for precise manipulation and task execution.
• Existing hand-eye and robot-world-hand-eye calibration methods have limitations in complex scenarios.

Purpose of the Study:

• To introduce two novel methods for robot-world-hand-eye calibration.
• To compare these methods against six state-of-the-art techniques.
• To analyze the impact of objective function choices (pose vs. reprojection error).

Main Methods:

• Exploration of two distinct geometrical interpretations: 'hand-eye' and 'robot-world-hand-eye'.
• Development of a robotic arm error modeling approach for realistic simulations.
• Testing on three real and three simulated datasets, including noisy conditions.

Main Results:

• Proposed methods demonstrate significant improvements in accuracy and robustness across various metrics.
• Comparative analysis shows superior performance against six leading state-of-the-art calibration methods.
• Robustness validated in both ideal and noisy simulated environments.

Conclusions:

• The novel methods offer enhanced performance for robot-world-hand-eye calibration.
• The study provides valuable insights into objective function selection for calibration.
• The proposed error modeling enhances the realism and reliability of simulated calibration data.