Action selection using priority for each task — Realization of effective processing by dynamical update of priority

In recent years, some research makes the robot treat multi-task. Importance for each task is used to do decision making and is usually changed according to the situation. We proposed method that does decision making according to a change of importance for each task. In this method, we defined priority as importance for task. The system updated priority per n action, the update frequency is fixed. But there is a risk that the system acts without calculating priority according to a situation because update frequency is fixed. In this paper, we solve a problem by dynamically changing update frequency of priority. To calculate priority according to a situation, we propose a method to decide update frequency for priority by using the process of calculating the priority. We carried out an experiment that set three tasks to the robot applied proposal technique. From experimental results, we confirmed the usefulness of proposal technique.