Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : May 10, 2017 - May 13, 2017
In this paper, we present a surveillance system based on an unmanned aerial vehicle, UAV, for poaching ships. In this surveillance system, the UAV is required to prevent the ships from arriving at their destination and intruding into the patrolling area. However, these two purposes cause a trade-off problem. For this challenge, we adopt a probabilistic approach. For unknow ships, first off, Bayesian leaning is used in order for the UAV to estimate the poaching trend of the ships. On the basis of the trend, value iteration is further used for optimizing the patrolling path to the destination. Through simulation experiments, we show that the UAV based on the proposed strategy is enabled to patrol the environment depending on the poaching ships.
