Evolutionary Creation of Adaptive Neurocontrollers Inspired by the Dynamically Rearranging Function in Biological Nervous Systems

受生物神经系统动态重排功能启发的自适应神经控制器的进化创造

• 批准号: 12650442
• 负责人: ISHIGURO Akio
• 金额: $ 1.86万
• 依托单位: Nagoya University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2000
• 资助国家: 日本
• 起止时间: 2000 至 2001
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-12650442/
• 关键词: Evolutionary Robotics; Neural Networks; Neuromodulation; Neuromodulators; Real-time Adaptation; Autonomous Mobile Robots; Locomotion Control; Central Pattern Generators; 適応能力; 自律律移動ロボット; 拡散・反応; 身体性

Recently the Evolutionary Robotics (ER) approach has been attracting a lot of concern in the field of robotics and artificial life. In contrast to the conventional approaches where designers have to construct controllers in a top-down manner, the methods employed in the ER approach have significant advantages since they can autonomously and efficiently construct controllers by taking embodiment (e.g. mechanical dynamics, physical size and shape of robots, sensor/motor properties and disposition etc.) and the interaction dynamics between the robot and its environment into account.In light of these facts, in this research project, an evolutionary creation of adaptive neurocontrollers has been investigated. To this end, the concept of neuromodulation, allowing to evolve neural networks which can adjust not only the synaptic weights but also the structure of the neural network by blocking and/or activating synapses or neurons according to the current situation has been introduced. This idea has been applied to the control of a peg-pushing robot and a legged robot, which is intrinsically unstable. Simulations have been conducted and the results have showed that this approach is remarkably promising.

近年来，进化机器人（ER）方法在机器人和人工生命领域引起了广泛的关注。与设计者必须以自上而下的方式构造控制器的传统方法相比，ER方法中采用的方法具有显著的优点，因为它们可以通过采用实施例（例如，机械动力学、机器人的物理尺寸和形状、传感器/电机特性和布置等）自主地且有效地构造控制器。以及机器人与环境之间的相互作用动力学。鉴于这些事实，在本研究项目中，研究了自适应神经控制器的进化创造。为此，已经引入了神经调制的概念，其允许进化神经网络，所述神经网络不仅可以通过根据当前情况阻断和/或激活突触或神经元来调整突触权重，而且还可以调整神经网络的结构。这一思想已被应用到一个peg-pushing机器人和腿的机器人，这是本质上不稳定的控制。仿真结果表明，这种方法是非常有前途的。

项目成果

A.Fujii et al.: "Evolving Bipedal Locomotion with a Dynamically-Rearranging Neural Network"Proc. of 6^<th> European Conference on Artificial Life. 509-518 (2001)

A.Fujii 等人：“利用动态重排神经网络进化双足运动”Proc。

R-Pfeiffer, C.Scheier 著(石黒章夫 他監訳): "知の創成 -身体性認知科学への招待-"共立出版. 724 (2001)

R-Pfeiffer, C.Scheier（由 Akio Ishiguro 等人监督翻译）：“知识的创造 - 具身认知科学的邀请”Kyoritsu Shuppan 724 (2001)。

A.Ishiguro et al.: "Evolving an Adaptive Controller for a Legged-Robot with Dynamically-Rearranging Neural Networks"Proc.of the 6th International Conference on The Simulation of Adaptive Behavior (SAB2000). 235-244 (2000)

A.Ishiguro 等人：“利用动态重排神经网络进化腿式机器人的自适应控制器”第六届自适应行为模拟国际会议 (SAB2000) 的会议记录。

藤井亮暢 他: "多型性を有するニューラルネットワークによる歩行の生成"ロボテス・メカトロニクス講演会2001講演論文集. (CD-ROM). (2001)

Akinobu Fujii 等人：“使用多态神经网络生成步态”机器人和机电一体化会议 2001 年论文集（CD-ROM）。

R. Pfeifer, C. Scheier著(石黒章夫 他監訳): "知の創成 -身体性認知科学への招待-"共立出版. 724 (2001)

R. Pfeifer、C. Scheier（由 Akio Ishiguro 等人监督翻译）：“知识的创造 - 具身认知科学的邀请”Kyoritsu Shuppan 724 (2001)。