Minimize Tracking Occlusion in Collaborative Pick-and-Place Tasks: An Analytical Approach for Non-Wrist-Partitioned...

Area of Science:

• Robotics and Automation
• Computer Vision
• Industrial Engineering

Background:

• Industrial pick-and-place applications require reliable visual tracking of parts.
• Collaborative assembly lines face productivity losses and failures due to recurrent occlusions caused by manipulator motion.
• Existing methods struggle with occlusions from both the manipulator and human operators.

Purpose of the Study:

• To develop a complete solution for maintaining an occlusion-free line of sight between a camera and a target object in collaborative assembly.
• To address occlusions caused by both the 6R manipulator and human operators.
• To enhance the efficiency and safety of collaborative assembly workstations.

Main Methods:

• Utilizing an actuated camera for object detection and operator tracking.
• Employing a complete set of solutions from the univariate polynomial equation for inverse kinematics (IK).
• Integrating collision and occlusion avoidance optimizations into an analytical-based method.

Main Results:

• The analytical IK approach provides multiple joint positions (postures) for each solution.
• The method successfully extracts optimal postures that minimize occlusion risks.
• Simulations and physical deployments on commercial hardware validate the approach's effectiveness.

Conclusions:

• The proposed analytical method offers a robust solution for occlusion-free visual tracking in industrial settings.
• This approach significantly enhances efficiency and safety in collaborative assembly workstations.
• The method provides a superior alternative to numerical iterative solving methods for IK in this context.