Anatomically Correct Testbed Hand Control: Muscle and Joint Control Strategies

<p>Human hands are capable of many dexterous grasping and manipulation tasks. To understand human levels of dexterity and to achieve it with robotic hands, we constructed an anatomically correct testbed (ACT) hand to investigate the biomechanical features and neural control strategies of the human hand. This paper focuses on developing control strategies for the index finger motion of the ACT Hand. A direct muscle position control and a force-optimized joint control are implemented as building blocks and tools for comparisons with future biological control approaches. We show how Gaussian process regression techniques can be used for nonlinear parameter estimation in both controllers. Our experiments demonstrate that the direct muscle position controller allows for accurate and fast position tracking, while the force-optimized joint controller allows for exploitation of actuation redundancy in the finger critical for this redundant system. Furthermore, a thogrough comparison between Gaussian processes and polynomials for non-linear regression shows that Gaussian processes provide significantly better parameter estimation and tracking performance. This first control investigation on the ACT hand opens doors to implement biological strategies observed in humans and achieve the ultimate human-level dexterity.</p>

http://www.cs.washington.edu/ai/Mobile_Robotics/postscripts/act-hand-icr... PDF