To understand human levels of dexterity and to achieve it with robotic hands, we constructed an anatomically correct testbed (ACT) hand. The ACT Hand allows for the investigation of the biomechanical features and neural control strategies of the human hand. This project focused 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. In addition, Gaussian processes and least squares regression are used in nonlinear parameter estimation in both controllers and performance was compared. The direct muscle position controller allowed for accurate and fast position tracking, while the force-optimized joint controller allowed for exploitation of actuation redundancy in the finger critical for this redundant system. Gaussian processes have shown to provide better parameter estimation and tracking with a cost of slower processing time. This first control investigation on the ACT hand opens doors to implement biological strategies observed in humans and achieve the ultimate human-level dexterity.