The Research On Route Design Based On Ant Colony Algorithm

With the constant progress of science and technology, vessel sailling is developping towards the intelligent direction. While it is realized that the vessel sails intelligently, the route design technology is undoubtedly one of the key technology. With the backround of the intelligent vessel sailing, the precondition of study of the new theory and method and the purpose of solving the criyical engineering application problems, this dissertation studies the path planning problem, environmental modeling and overall route search in route design problem deeply. The main work of the dissertation is as follows:Firstly, the dissertation has summarized the basic content and characteristic of the path planning problem, and analysised environmental modeling and route of planning problems. on Carried on comparative analysis to several methods of environmental modeling, at the same time, the deep research was made on relatively several kinds of intelligent algorithms which are often used in path planning. Secondly, further research on basic algorithms and the optimizing of principle of ant was made. Having analyzed the impact of ant colony algorithm on performance of main parameters, and made the emulation with MATLAB.The dissertation has summarized the characteristic of route design problem, analysised the difference between TSP and route design problem. In order to settle the contradictory between convergence speed and precocity in ant colony algorithm,a new thought was introduced.the thought can improve the algorithm capability by dynamically adjusting the information updating method and the selected probabilities.the algorithm can keep good balance between accelerating convergence and averting precocity and stagnation. The route design problem is deeply analysed,to the characteristic of the vessel sailing environment,The environment model has been established,by using the MAKLINK, and adopted ant colony algorithm as the route design algorithm. The algorithm which was applied to the route design problem has been designed detailedly. To judge the route by defining the think function. The smooth method has been proposed to make the route more smooth, and make the route approach the true route even more. To set up the environmental model of ocean by using the MAKLINK law finally, An adaptive ant colony algorithm based on the equiliblium of the ant distribution was presented and the overall searching model based on ant colony algorithm has been proposed, the resoult of the experiment show that the improved measure of the paper have better convergence and dependability.The algorithm which was adopted by the dissertation has received the ideal result with emulation, while the convergence property, validity and commonability have been verified. the progress of the algorithm has been verified through the convergence property difference. The dissertation compared the algorithm with basic ant colony algorithm by simulation.