Reinforcement Learning for Hand Grasp with Surface Multi-field Neuroprostheses
Hand grasp is a complex system that plays an important role in the activities of daily living. Upper-limb neuroprostheses aim at restoring lost reaching and grasping functions on people suffering from neural disorders. However, the dimensionality and complexity of the upper-limb makes the neuroprostheses modeling and control challenging. In this work we present preliminary results for checking the feasibility of using a reinforcement learning (RL) approach for achieving grasp functions with a surface multi-field neuroprosthesis for grasping. Grasps from 20 healthy subjects were recorded to build a reference for the RL system and then two different award strategies were tested on simulations based on neuro-fuzzy models of hemiplegic patients. These first results suggest that RL might be a possible solution for obtaining grasp function by means of multi-field neuroprostheses in the near future.
KeywordsNeuroprostheses Functional electrical stimulation Grasp Reinforcement learning Modeling and control
Authors would like to thank to Intelligent Control Research Group of UPV/EHU for giving the means of carrying out this work and to the Health Division of Tecnalia for its continuous support.
- 18.Gillen, G.: Stroke Rehabilitation: A Function-Based Approach. Elsevier Health Sciences, St. Louis (2015)Google Scholar
- 19.Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT press, Cambridge (1998)Google Scholar
- 21.Imatz-Ojanguren, E., Irigoyen, E., Valencia-Blanco, D., Keller, T.: Neuro-fuzzy models for hand movements induced by functional electrical stimulation in able-bodied and hemiplegic subjects. Med. Eng. Phys. (2016). (In Press)Google Scholar