A Personal Navigation System for Sightseeing across Multiple days

Recently, various personal navigation systems are available on the market. In this paper, we propose a new scheduling function to compose sightseeing tours across multiple days. Scheduling for multiple days is challenging since the number of possible sequences to travel destinations becomes huge and accommodation places have to be carefully chosen considering the schedules before and after each stay. So, some techniques are required to calculate the sightseeing schedule across multiple days at practically short time. In our method, we divide the target area into several sub-areas, and we restrict each day trip to be in a limited number of areas with infrequent passing of area boundary, in order to decrease the number of sequences. We have designed and implemented a genetic algorithm to solve this scheduling problem at a practical time. In this algorithm, accommodation places are represented in chromosome as exclusive alleles, to make schedules which contain only one accommodation place for each day. In order to evaluate our method, we made various sightseeing schedules across multiple days using the digital map of the Tohoku area (Japan Geographical Survey Institute numeric map 25000), and confirmed that the proposed algorithm could compute the near best 6 days schedule in several tens of seconds.