Dynamic Path Planning Based on Adaptable Ant Colony Optimization Algorithm

• Main Authors: Shih-Shih Chi, 紀詩詩
• Other Authors: Kun-Ming Yu
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/6xp5k2

碩士 === 中華大學 === 資訊工程學系 === 105 === In recent years, many tourist attractions have suffered heavy traffic problems during vacations or holidays due to the convenience of transportation and the growing popularity of leisure activities. Traffic is overwhelming in attraction areas during almost every rush hour, as many cars converge from all directions to cause traffic jams. However, according to common experience, different recreational attractions are not necessary located along a single route in a given tourist area. If all the recreational attractions in a given area are arranged in sequence based on the order visited, visitors may suffer schedule delays due to traffic problems occurring in early spots of this sequence, as tourists do not always visit attractions in the same order. This thesis presents an Adaptable Ant Colony Optimization Algorithm(AACO) to solve the traffic jam problems. This algorithm can determine the priority of visitation to different attractions, using travel time and the distance between two attractions to determine the optimal path arrangement taken by each visitor. Every attraction may only be passed through once. The method is then implemented again to determine the next attraction, and so forth, until all attractions have been visited. This thesis combines a consideration of leisure itineraries and the practical operation of the route planning model in order to establish a dynamic planning system that instantaneously provides optimal route information. The goal is to create a system that considers visitor preferences, and uses Google Maps API to assist in tailoring travel routes to the individual consumer. The experimental results showed that this research can plan the route according to the desire degree of tourists to scenic spots in a short period. The sum of the desired values calculated by this algorithm is greater than that of the original ant optimization algorithm and the improved ant optimization algorithm. It is because the route selected is calculated based on the attraction desire priority. The results show that the mean total desired value of improved ant algorithm is 25.53% higher than that of the traditional ant optimization algorithm, the average travel time is reduced by 34.32%, and the average computing time is reduced by 25.53%. It can be seen from the data that the algorithm can effectively improve the travel route planning problem.