Generative GaitNet
This work addresses predictive gait simulation for biomechanics and medical applications, but appears incremental as it builds on existing deep reinforcement learning and physics-based simulation methods.
The paper tackled the problem of simulating human gait by developing Generative GaitNet, a deep reinforcement learning network that controls a full-body musculoskeletal model with 304 musculotendons, and demonstrated its ability to generate healthy and pathologic gaits in real-time physics-based simulation.
Understanding the relation between anatomy andgait is key to successful predictive gait simulation. Inthis paper, we present Generative GaitNet, which isa novel network architecture based on deep reinforce-ment learning for controlling a comprehensive, full-body, musculoskeletal model with 304 Hill-type mus-culotendons. The Generative Gait is a pre-trained, in-tegrated system of artificial neural networks learnedin a 618-dimensional continuous domain of anatomyconditions (e.g., mass distribution, body proportion,bone deformity, and muscle deficits) and gait condi-tions (e.g., stride and cadence). The pre-trained Gait-Net takes anatomy and gait conditions as input andgenerates a series of gait cycles appropriate to theconditions through physics-based simulation. We willdemonstrate the efficacy and expressive power of Gen-erative GaitNet to generate a variety of healthy andpathologic human gaits in real-time physics-based sim-ulation.