Bio-inspired tendon-driven systems: Computational analysis, optimization, and hardware implementation

This thesis work focuses on the design and optimization of tendon-driven systems. One of the chief reasons these systems can be termed "bio-inspired" is due to the fact that they produce forces and movements via tendons which are connected to actuators and have uni-directional action (they can only pull, not push). In particular, the human hand sets itself apart from the rest of the body's neuromuscular systems in that there are no muscles in the fingers themselves: all the muscles are proximal to the fingers. The study and analysis of tendon-driven hands and fingers suggest that i) the human hand is mechanically optimized for grasping capabilities and ii) use of bio-inspired principles in robotic systems can drastically improve force-production capabilities, grasp strength, and stiffness control performance. Some of the bio-inspired principles which are proven throughout this work to be very beneficial in robotic systems are asymmetry and skewed combinations of design parameters. This work not only has strong implications for the design of commercial robotic and prosthetic hands, but also broadly seeks to inspire creativity in design and optimization routines for complex problems in many different engineering disciplines.