Novel affordable light-weight underactuated biomimetic multi-DOF prosthetic hands

The overarching aim is to apply theoretical and experimental approaches to better understand the mechanism requirements and then develop highly novel solutions for multi-DOF prosthetic hands. In this context, to minimise the number of independent actuators required, we will adopt novel approaches to mechanism reconfiguration (for different grip patterns) and differential kinematics (to conform to different object shapes). Digital additive manufacturing approaches will also be applied in this project. To achieve this we envisage the following work: a) Review literature on the grip patterns actually used in practise and the associated bimanual tasks that matter most for amputees. Many of the grip patterns provided by existing multi-DOF hands are rarely used (Resnik, L., Acluche, F., and Borgia, M., 2018). In this way, a sensible set of grip patterns will be identified. b) Based on this set of grip patterns, a formal kinematic specification will be established. This will be formulated mathematically so that it can be used as input for theoretical mechanism synthesis methods. c) Apply formal mechanism design methods (based on mechanism theory) to generate alternative hybrid soft-rigid mechanism designs to achieve: i) mechanism reconfiguration for different grip patterns; ii) differential kinematics to conform to different object shapes; and iii) intelligent adaptation to grip force. d) Use virtual prototyping to model alternative multi-DOF hand designs incorporating our novel mechanisms and then to narrow these down to 1 or 2 preferred designs. e) Produce research prototypes via Salford's Maker Space and NERIC (North England Robotics Innovation Centre), and test with a small cohort of amputees in both the UK and China.