Sim-to-Real Transfer for Muscle-Actuated Robots via Generalized Actuator Networks
2026-04-10 • Robotics
RoboticsMachine Learning
AI summaryⓘ
The authors worked on improving robots that use soft muscles and tendons, which are tricky to control because of their complex behavior like friction and delays. They created a new method called Generalized Actuator Network (GeAN), which learns how these muscles act from simple movement data instead of needing difficult measurements. By combining this with existing simulation tools, they trained robot arm movements entirely in simulation and successfully transferred those skills to a real robot. This is the first time such a method has worked for a robot arm with four moving parts powered by artificial muscles.
tendon drivessoft muscle actuationnonlinearitieshysteresissim-to-real transferneural network modelrigid body simulationpneumatic artificial musclesrobot arm controlpolicy learning
Authors
Jan Schneider, Mridul Mahajan, Le Chen, Simon Guist, Bernhard Schölkopf, Ingmar Posner, Dieter Büchler
Abstract
Tendon drives paired with soft muscle actuation enable faster and safer robots while potentially accelerating skill acquisition. Still, these systems are rarely used in practice due to inherent nonlinearities, friction, and hysteresis, which complicate modeling and control. So far, these challenges have hindered policy transfer from simulation to real systems. To bridge this gap, we propose a sim-to-real pipeline that learns a neural network model of this complex actuation and leverages established rigid body simulation for the arm dynamics and interactions with the environment. Our method, called Generalized Actuator Network (GeAN), enables actuation model identification across a wide range of robots by learning directly from joint position trajectories rather than requiring torque sensors. Using GeAN on PAMY2, a tendon-driven robot powered by pneumatic artificial muscles, we successfully deploy precise goal-reaching and dynamic ball-in-a-cup policies trained entirely in simulation. To the best of our knowledge, this result constitutes the first successful sim-to-real transfer for a four-degrees-of-freedom muscle-actuated robot arm.