Design and Empirical Investigation of a Biomimetic Anthropomorphic Robotic Arm Featuring Human-Like Biomechanical Capabilities and Performance

Abstract

In the modern technological paradigm, robotics has found applications from manufacturing to healthcare. Current robots grapple with challenges in safety, mobility, compactness, and load-bearing. This research delves into the human arm anatomy, examining bones, ligaments, tendons, and muscles, to pinpoint design deficiencies in current humanoid robots. Drawing on these anatomical insights, a bionic robotic arm that addresses these identified limitations is proposed and validated. Central to this study are three compliant actuators: ICA, ECA, and MCA. Notably, the MCA's nonlinear stiffness closely mirrors biological joint dynamics. Meanwhile, the ECA and ICA boast high power ratios, and the MCA displays a broad spectrum of joint stiffness. The proposed robotic forearm and elbow design aligns closely with human skeletal biomechanics. Moving beyond the conventional emphasis on soft tissues, the proposed design emulates biological joints, ensuring that the robotic arm matches the human range of motion and torque capacities. Additionally, the research delves into the biomechanics of the human shoulder. It addresses current design inadequacies and presents a biomimetic glenohumeral joint that embodies key properties of the biological shoulder, especially stability and mobility. Tests confirm the enhanced flexibility and load-bearing capabilities of this design. In the ultimate evaluation of the robotic arm, the study concludes that anatomically replicating soft tissues, like ligaments, joint capsules, cartilage, and tendons, substantially enhances the joint stability and flexibility of the bionic robot.

Date of Award	29 Aug 2023
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Jyoti Sinha (Main Supervisor)