A Structural Design and Workspace Analysis of a Lumbar Massage Robotic Arm

Background: In modern rehabilitation medicine, the introduction of robotic technology provides patients with more precise and personalized rehabilitation services, especially demonstrating significant application potential in functional recovery training. Objective: To address the needs for compliance, lightweight design, and safety in rehabilitation robots, this study proposes a hybrid-driven lumbar massage robotic arm based on motors, cylinders, and pneumatic artificial muscles, and improves workspace calculation accuracy. Methods: Through structural design, combining a hybrid drive of motors, cylinders, and pneumatic artificial muscles, the robotic arm achieves optimal workspace and compliance. This design holds patent potential. The improved D-H method establishes the robotic arm's kinematic model, and an enhanced Monte Carlo method analyzes its workspace. Results: The effect of each parameter on workspace calculation accuracy was investigated and compared to the traditional Monte Carlo method. Results indicate that the improved Monte Carlo method maintains high accuracy with 68% fewer point counts than the traditional method. The error rate along each coordinate axis is within 0.2%, with an average error rate of 0.1%, significantly lower than the traditional Monte Carlo method’s 1.02%. Additionally, at a loop count of 3, the improved Monte Carlo method’s average error rate of 0.12% is much lower than the traditional method’s 1.34%. Conclusion: The improved Monte Carlo method offers higher accuracy while maintaining computational efficiency, outperforming the traditional Monte Carlo method and laying the groundwork for further optimization and control of the robotic arm.