Reinforcement learning for non-prehensile manipulation: Transfer from simulation to physical system
This addresses the challenge of sample efficiency and safety in robot control for researchers in robotics and reinforcement learning, though it is incremental as it builds on existing model-based methods.
The paper tackled the problem of transferring reinforcement learning policies from simulation to a physical robot system for non-prehensile manipulation, showing that policies trained entirely in simulation successfully pushed an object to target positions on three Phantom robots without additional real-world training.
Reinforcement learning has emerged as a promising methodology for training robot controllers. However, most results have been limited to simulation due to the need for a large number of samples and the lack of automated-yet-safe data collection methods. Model-based reinforcement learning methods provide an avenue to circumvent these challenges, but the traditional concern has been the mismatch between the simulator and the real world. Here, we show that control policies learned in simulation can successfully transfer to a physical system, composed of three Phantom robots pushing an object to various desired target positions. We use a modified form of the natural policy gradient algorithm for learning, applied to a carefully identified simulation model. The resulting policies, trained entirely in simulation, work well on the physical system without additional training. In addition, we show that training with an ensemble of models makes the learned policies more robust to modeling errors, thus compensating for difficulties in system identification.