Autonomous decision making by the self-generated priority under multi-task

In recent years, a robot is required to perform multitask autonomously in human living space. It needs to take actions according to situations. We proposed a method which does decision making on a robot with multi-task according to a situation by using an importance of each task. To respond to changes in importance of task, the robot learned each task independently by using reinforcement learning. An action is selected uniquely using action values and importance of each task in this system. The parameters are designed according to the value that represents the status of each task as an index for evaluating the importance. Therefore, it is necessary to design parameters suitable for the environment for each task. If the environment changes, the parameters must be designed accordingly. Therefore, in this research, we propose an autonomous decision-making method based on priority self-generation. The robot self-generates the priority of each task based on the experience gained by the robot, and realizes an action selection system with the priority matching the environment. We carried out an experiment which set three tasks to the robot applied proposal method. From experimental results, we confirmed the usefulness of proposed method.