Biomechanic Posture Stabilisation via Iterative Training of Multi-policy Deep Reinforcement Learning Agents
This addresses posture stabilization for biomechanical applications, representing an incremental improvement in training methods for deep reinforcement learning agents.
The paper tackled the error propagation problem in training an AI agent to maintain a stable standing posture for a digital musculoskeletal avatar, achieving an increase in standing duration from 4 seconds to 348 seconds and generalizing to noise for 108 seconds.
It is not until we become senior citizens do we recognise how much we took maintaining a simple standing posture for granted. It is truly fascinating to observe the magnitude of control the human brain exercises, in real time, to activate and deactivate the lower body muscles and solve a multi-link 3D inverted pendulum problem in order to maintain a stable standing posture. This realisation is even more apparent when training an artificial intelligence (AI) agent to maintain a standing posture of a digital musculoskeletal avatar due to the error propagation problem. In this work we address the error propagation problem by introducing an iterative training procedure for deep reinforcement learning which allows the agent to learn a finite set of actions and how to coordinate between them in order to achieve a stable standing posture. The proposed training approach allowed the agent to increase standing duration from 4 seconds using the traditional training method to 348 seconds using the proposed method. The proposed training method allowed the agent to generalise and accommodate perception and actuation noise for almost 108 seconds.