Reinforcement learning of electric prosthetic hand with depth camera in simulation environment

Takuya Ayukawa, Nobuhiko Yamaguchi, Osamu Fukuda, Hiroshi Okumura


In this paper, we propose a control scheme for electric prosthetic hand system based on depth camera images. We develop a simulation environment of an electric prosthetic hand based on depth camera images by Unity and acquire control scheme by reinforcement learning in the simulation environment. The performance of the proposed method is assessed in experiments with stick picking task. The experiment highlights the electric prosthetic hand based on depth camera produces higher cumulative reward and outperforms the conventional electric prosthetic hand based on 2D camera images.


Electric prosthetic hand; Reinforcement learning; Depth camera; Unity

Full Text:



  • There are currently no refbacks.