Deep Reinforcement Learning Control of an Avatar and Exoskeleton
We present a novel approach for controlling a unilateral lower extremity exoskeleton with deep reinforcement learning utilizing a Long-term Short-term Memory network with a Proximal Policy Optimization Model. We create an avatar in Unity to reflect the movement while also controlling our exoskeleton.
Authors
- B. Gonzalez
- Robert Dizor (ORCID: https://orcid.org/0000-0001-6127-2868)
- Anil Raj (ORCID: https://orcid.org/0000-0002-3131-7268)
Institutions
- Florida Institute for Human and Machine Cognition (US)
- University of West Florida (US)
Publication Details
- Published
- 2024-08-14
- DOI
- https://doi.org/10.1109/rapid60772.2024.10647027
- Citations
- 1
- Primary Topic
- Muscle activation and electromyography studies
- Type
- article
- Field-Weighted Citation Impact
- 0.18
Funders
- Embry-Riddle Aeronautical University
- University of West Florida