A compensation strategy of end-effector pose precision based on the virtual constraints for serial robots with RDOFs

Area of Science:

• Robotics
• Control Systems Engineering

Background:

• Serial robots with redundant degrees-of-freedom require high end-effector pose precision for complex tasks.
• Cumulative errors in joint actuators present a significant challenge to achieving this precision.

Purpose of the Study:

• To propose and validate a novel virtual-constraints-based end-effector pose compensator for redundant serial robots.
• To address and mitigate end-effector pose errors caused by joint actuator inaccuracies.

Main Methods:

• The proposed method treats actual joint angles as virtual constraints.
• These virtual constraints, along with the desired end-effector pose, determine compensation angles for other joints.
• Compensation angles are applied to adjust other joint angles, correcting pose errors.

Main Results:

• Experimental validation on a 3-degree-of-freedom planar serial robot demonstrated significant error reduction.
• The compensator achieved a maximum reduction in foot end position error exceeding 75%.
• The strategy enhances end-effector pose precision, complementing existing high-precision joint control methods.

Conclusions:

• The virtual-constraints-based compensator is an effective strategy for improving the end-effector pose precision of redundant serial robots.
• This approach offers a novel solution with experimental evidence for compensating pose errors in complex robotic systems.