Modelling Generalized Forces with Reinforcement Learning for Sim-to-Real Transfer

Abstract

Learning robotic control policies in the real world gives rise to challenges in data efficiency, safety, and controlling the initial condition of the system. On the other hand, simulations are a useful alternative as they provide an abundant source of data without the restrictions of the real world. Unfortunately, simulations often fail to accurately model complex real-world phenomena. Traditional system identification techniques are limited in expressiveness by the analytical model parameters, and usually are not sufficient to capture such phenomena. In this paper we propose a general framework for improving the analytical model by optimizing state dependent generalized forces. State dependent generalized forces are expressive enough to model constraints in the equations of motion, while maintaining a clear physical meaning and intuition. We use reinforcement learning to efficiently optimize the mapping from states to generalized forces over a discounted infinite horizon. We show that using only minutes of real world data improves the sim-to-real control policy transfer. We demonstrate the feasibility of our approach by validating it on a nonprehensile manipulation task on the Sawyer robot.

References in corpus (6)

• Learning agile and dynamic motor skills for legged robots
• End-to-End Training of Deep Visuomotor Policies
• Emergence of Locomotion Behaviours in Rich Environments
• IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
• Model-Based Reinforcement Learning via Meta-Policy Optimization
• Reinforcement and Imitation Learning for Diverse Visuomotor Skills

Cited by in corpus (6)

• Never Stop Learning: The Effectiveness of Fine-Tuning in Robotic Reinforcement Learning
• Gym-Ignition: Reproducible Robotic Simulations for Reinforcement Learning
• SimGAN: Hybrid Simulator Identification for Domain Adaptation via Adversarial Reinforcement Learning
• A Simple Approach to Continual Learning by Transferring Skill Parameters
• DROID: Minimizing the Reality Gap using Single-Shot Human Demonstration
• "What, not how": Solving an under-actuated insertion task from scratch