This paper carried out a research of the dynamic route-planning of the ship. Using Memetic algorithm and with collision regulars on the sea, the format of ship-encounter is analyzed and furthermore, the ways to evade is determined. Finally, the constraints of the obstruction cost and the length cost, the dynamic path planning is conducted. According to the results of the path planning, the ship simulation manipulator is used to simulate and verify the two situations including crossing and head-to-head, and then analyze the path planning process. From the perspective of maritime practice, the empirical results of avoidance are used as a comparison to verify the validity and applicability of the algorithm.
With the development of the world economy, the "One Belt, One Road" initiative and the "Ocean Power" strategy, the marine economy has become a new economic growth point. The development of the marine economy has reached an unprecedented height. With the full implementation of the global E-navigation strategy, navigation is shifting towards to "intelligent maneuvering, intelligent decision-making, unmanned driving, and wide-scale navigation protection". The era of unmanned cargo ships has arrived. The technology mentioned above requires high-end scientific theoretical methods for support and highly intelligent navigation technology as a guarantee to achieve unmanned driving. At the same time, maritime tasks are becoming more frequent, so issues such as maritime safety and efficiency of offshore operations have received much attention. In the face of complex time, space constraints and sudden task sets, more situations require multiple ships to work together to maximize efficiency.
In the case of ship path planning, Chang(Chang K Y, Jan G E. Parberry I.2003) designed a novel calculation model, which uses the maze routing algorithm to achieve accurate analysis of the collision avoidance route that may occur in the event of an accident. Ito M(Ito M, Zhang F, Yoshida N.1999) adopted the concept of the security domain to create and simulate the space, and then determine the optimal path adjustment algorithm through four different and meaningful parameters. Yang and Gao (Yang Ke. 2015)(Gao RJ. 2016) implemented a more accurate path planning by establishing a model of the turbulent region of the pier and using hybrid intelligent algorithms under multiple constraints such as steering angle and navigable water boundary. The study provides a scientific basis and theoretical basis for decreasing bridge impact accidents. Based on the characteristics of motion, Wu (Wu Bo, Wen YQ, Xiao CS.2013) explored the autonomous collision avoidance algorithm of unmanned boats when performing tasks in relatively harsh environments. Xiang (Xiang ZQ, Zhai Chao, Du KJ, et al.2015) used the heuristic algorithm to initialize the particles and used the path smoothing optimization method, then he integrated the maritime rules to realize the static and dynamic path planning of the surface unmanned boat.
