実ロボットによる行動学習のための状態空間の漸次的構成

抄録

Reinforcement learning has recently been receiving increased attention as a method for robot learning with little or no a priori knowledge and higher capability of reactive and adaptive behaviors. However, there are two major problems in applying it to real robot tasks: how to construct the state space, and how to accelerate the learning. This paper presents a method by which a robot learns a purposive behavior within less learning time by incrementally segmenting the sensor space based on the experiences of the robot. The incremental segmentation is performed by constructing local models in the state space, which is based on the function approximation in terms of the sensor outputs and the reinforcement signal to reduce the learning time. The method is applied to a soccer robot which tries to shoot a ball into a goal. The experiments with computer simulations and a real robot are shown. As a result, our real robot has learned a shooting behavior within less than one hour training by incrementally segmenting the state space.