Sim-to-Real: Learning Agile Locomotion For Quadruped Robots
This addresses the challenge of reducing manual tuning for robot locomotion, though it is incremental by building on existing sim-to-real methods.
The paper tackles the problem of automating agile locomotion for quadruped robots by using deep reinforcement learning in simulation and transferring policies to real robots, achieving successful real-world performance for trotting and galloping gaits.
Designing agile locomotion for quadruped robots often requires extensive expertise and tedious manual tuning. In this paper, we present a system to automate this process by leveraging deep reinforcement learning techniques. Our system can learn quadruped locomotion from scratch using simple reward signals. In addition, users can provide an open loop reference to guide the learning process when more control over the learned gait is needed. The control policies are learned in a physics simulator and then deployed on real robots. In robotics, policies trained in simulation often do not transfer to the real world. We narrow this reality gap by improving the physics simulator and learning robust policies. We improve the simulation using system identification, developing an accurate actuator model and simulating latency. We learn robust controllers by randomizing the physical environments, adding perturbations and designing a compact observation space. We evaluate our system on two agile locomotion gaits: trotting and galloping. After learning in simulation, a quadruped robot can successfully perform both gaits in the real world.