仮想方策勾配を用いた強化学習のハイブリッド手法による倒立振子の平衡制御
A Hybrid Approach for Reinforcement Learning Using Virtual Policy Gradient for Balancing an Inverted Pendulum
シミュレーション上で方策勾配法により倒立振子の平衡制御を学習し、その重みを実機に転移することで、高速かつ堅牢な制御を実現した。
著者: Dylan Bates
分類: cs.LG, cs.RO
原文アブストラクト
Using the policy gradient algorithm, we train a single-hidden-layer neural network to balance a physically accurate simulation of a single inverted pendulum. The trained weights and biases can then be transferred to a physical agent, where they are robust enough to to balance a real inverted pendulum. This hybrid approach of training a simulation allows thousands of trial runs to be completed orders of magnitude faster than would be possible in the real world, resulting in greatly reduced training time and more iterations, producing a more robust model. When compared with existing reinforcement learning methods, the resulting control is smoother, learned faster, and able to withstand forced disturbances.