vix.ing · top · new · best · stats

Sim-to-Real: Learning Agile Locomotion For Quadruped Robots

2018/04/27 by Jie Tan, Tan, Jie, Tingnan Zhang +14 · 114 citations
Computer Science · Engineering · #Agile software development #Artificial intelligence #Computer science #Control engineering #Engineering #Gait #Human–computer interaction #Process (computing) #Reinforcement Learning in Robotics #Reinforcement learning #Robot #Robot Manipulation and Learning #Robotic Locomotion and Control #Robotics #Simulation #Software engineering #cs.AI #cs.RO

paper · pdf · doi:10.48550/arxiv.1804.10332

published in arXiv (Cornell University) (Cornell University) · Accompanying video: https://www.youtube.com/watch?v=lUZUr7jxoqM

openalex publication_date 2018/04/27 · arxiv created 2018/05/16 · arxiv updated 2018/05/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/08

Abstract

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.

Citations

Cited by

Related