Neuromodulated Control of Bipedal Locomotion Using a Polymorphic CPG Circuit
Neuromodulated Control of Bipedal Locomotion Using a Polymorphic CPG Circuit
复制标题
使用多态 CPG 电路对双足运动进行神经调节控制
作者：
A. Ishiguro;Akinobu Fujii;P. E. Hotz
迄今为止，已经提出了使用神经回路或所谓的中央模式生成器（CPG）概念的各种方法来为腿式机器人创建敏捷运动。与这些方法相反，在本文中，我们提出了一种多态神经电路，可以根据当前情况实时动态改变其属性。为此，引入了神经调节的概念。为了验证该方法的可行性，将该概念应用于本质上不稳定的三维双足机器人的控制。通过对双足在不平坦地形上行走的模拟说明了自适应控制器的重要性，结果表明双足机器人通过动态改变施加到关节的扭矩输出成功地应对了环境扰动。此外，所提出的方法优于具有感觉反馈的整体 CPG 模型。
To date, various methods using the concept of neural circuit or so-called central pattern generators (CPGs) have been proposed to create agile locomotion for legged robots. In contrast to these approaches, in this article we propose a polymorphic neural circuit that allows the dynamic change of its properties according to the current situation in real time to be employed instead. To this end, the concept of neuromodulation is introduced. To verify the feasibility of this approach, this concept is applied to the control of a three-dimensional biped robot that is intrinsically unstable. The importance of an adaptive controller is illustrated with the simulations of biped walking on uneven terrain, and the results show that the biped robot successfully copes with environmental perturbation by dynamically changing the torque outputs applied to the joints. Furthermore, the proposed approach outperforms a monolithic CPG model with sensory feedback.