Watanabe Toshiki, Kubo Akihiro, Tsunoda Kai, Matsuba Tatsuya, Akatsuka Shintaro, Noda Yukihiro, Kioka Hiroaki, Izawa Jin, Ishii Shin, Nakamura Yutaka
Kyoto University, Kyoto, Japan.
Advanced Telecommunications Research Institute, Kyoto, Japan.
Sci Rep. 2025 Apr 2;15(1):11262. doi: 10.1038/s41598-025-94163-2.
We present a data-driven deep reinforcement learning (DRL) method for the optimization of a hierarchically structured control policy that includes the central pattern generator. This method, which is as a whole referred to as the hierarchical reinforcement learning with the central pattern generator (HRL-CPG), is then evaluated with the expectation of its applicability in real robot controls. We observed that stable gait motions were gained in a reasonably small number of trials and errors. Thus, it can be deduced that our HRL-CPG can be a candidate DRL method that enables dynamical systems such as real or realistic robots to adapt to a variety of environments within a moderate physical time.
我们提出了一种数据驱动的深度强化学习（DRL）方法，用于优化包含中枢模式发生器的分层结构控制策略。这种方法整体上被称为带有中枢模式发生器的分层强化学习（HRL-CPG），随后对其在实际机器人控制中的适用性进行了评估。我们观察到，在合理数量的试验和错误中获得了稳定的步态运动。因此，可以推断出我们的HRL-CPG可以成为一种候选的DRL方法，使诸如真实或逼真机器人等动态系统能够在适度的物理时间内适应各种环境。
