Flexible muscle-based locomotion for bipedal creatures

TLDR

We present a muscle‑based control method for simulated bipeds that optimizes muscle routing and control parameters. The method uses 3D simulated muscles to generate actuation forces, incorporates a neural‑delay model for all feedback paths, and optimizes muscle routing and control parameters. The controllers produce generic locomotion for diverse bipedal creatures, generate torque patterns respecting biomechanical constraints, and adaptively select gaits for speed, uneven terrain, perturbations, and steering.

Abstract

We present a muscle-based control method for simulated bipeds in which both the muscle routing and control parameters are optimized. This yields a generic locomotion control method that supports a variety of bipedal creatures. All actuation forces are the result of 3D simulated muscles, and a model of neural delay is included for all feedback paths. As a result, our controllers generate torque patterns that incorporate biomechanical constraints. The synthesized controllers find different gaits based on target speed, can cope with uneven terrain and external perturbations, and can steer to target directions.