System-software phantom for tremor suppression by using reinforcement learning

The code contains the environments for the reinforcement learning model. The first environment works with a physical model. The second environment works with a virtual model written on the base of the OpenSim.

• Python 3.7
• OpenAI Gym - open-source library for making gyms
• pySerial - library for communication between Arduino and Python
• Stable Baselines - library with implemented reinforcement learning algorithms

In order to use the environment, you should have

• Arduino Uno x2
• Sensor GY-61
• DC-motor x3 (connecting via L298N)

The General scheme is shown in Figure. The experimental setup consists of a system for suppression, a system for creating oscillations, and an accelerometer for evaluating the system's state.

General scheme

Important parameter

• Python 3.7
• OpenAI Gym - open-source library for making gyms
• OpenSim 4.2 - software for developing models of musculoskeletal structures
• Stable Baselines - library with implemented reinforcement learning algorithms

If you use this package in your publications or in other work, please cite it as follows:

@misc{rl_tremor_sup,
    title={System-software phantom for tremor suppression by using reinforcement learning},
    author={Anna Anikina},
    year={2018},
    publisher = {GitHub},
    journal = {GitHub repository},
    howpublished = {\url{https://github.com/cviaai/RL-TREMOR-SUPPRESSION/}}
    }

Anna @annshorn