Syuichi Yokoyama

Autonomous reconfiguration of robot shape by using Q-learning

A modular robot can be built with a shape and function that matches the working environment. We developed a four-arm modular robot system which can be configured in a planar structure. A learning mechanism is incorporated in each module constituting the robot. We aim to control the overall shape of the robot by an accumulation of the autonomous actions resulting from the individual learning functions. Considering that the overall shape of a modular robot depends on the learning conditions in each module, this control method can be treated as a dispersion control learning method. The learning object is cooperative motion between adjacent modules. The learning process proceeds based on Q-learning by trial and error. We confirmed the effectiveness of the proposed technique by computer simulation.

Man-machine interface for modular robot systems

A modular robot is composed of multiple modules, each comprising a sensor, an actuator, and a control system. Each module accumulates information about its own sensor, actuator, and connections to other modules, as well as communication information between adjoining modules. The user obtains this information via an interface, and can thus recognize the state of the robot and issue commands. However, when the number of modules becomes large, the amount of information sent from the modules becomes too much for the user to deal with effectively. Naturally, it also becomes more difficult for the user to issue commands to the modular robot as the number of modules increases. In this study, we developed an interface to present, in a simple manner, information aggregated in a certain module from other modules, and we examined its effectiveness in a modular robot composed of these modules.