Study On Ship Dynamic Path Planning In Multi-bridge Waters Area

With the rapid development of the transportation,the proposal for the national great development strategy of “the Belt and Road Initiatives” and “the Yangtze River Economic Zone”,the cluster construction of cross-river bridges has been formed in the Yangtze River,Pearl River and other waters,Which has been promoting the economic development of the river basin at the same time,has also brought new challenges to the ship safety and the bridge itself in those water areas.Wuhan Nanjing Yangtze River Bridge and Yangtze River Bridge have been collided many times,highlighting the necessity of researching the ship navigation safety problem in multi-bridges water area.Ship path planning is an important part in the ship autonomous navigation research field.In order to reduce the bridge-ship collision accidents,and to ensure the safety of bridge and ship,and to realize the ship autonomous navigation finally.Taking the wuhan section of the Yangtze River as an example,the ship navigation environment model and path planning cost model are established in multi-bridges water area.The study of ship path planning in multi-bridges water area are carried out by using existing intelligent algorithms under multiple constraints.The proposed algorithms can not only solve ship path planning problems,but also provide a theoretical basis for the autonomous navigation of ship or the USV(unmanned surface vessel)in other complex waters.According to analyzing the characteristics of multi-bridges water area in Yangtze River Wuhan section,the two-dimensional coordinate calculation formulas of pier center and each endpoint are proposed based on multi-bridges distribution,the turbulent region and safety margin around pier are defined.The bridge pier model with strong applicability to reflect the characteristics of the multi-bridges water area in Wuhan section is established.Analyzing three constraints including the navigable waters boundary,ship steering angle and the numbers of steering point.By establishing path length cost calculation model and obstacles navigation cost calculation model,using the weighted method of multi-objectives optimization method to get the path planning total cost calculation model,then the path planning problem is transformed into solving multi-dimensional nonlinear function minimum problem.Based on five algorithms including Dijkstra algorithm,a hybrid Ant Colony optimization based on Dijkstra algorithm,GA(Genetic Algorithm),PSO(Particle Swarm Algorithm),a hybrid intelligent algorithm by mixing PSO and GA,using matlab software,the ship path planning simulations are carried out under static environment in multi-bridges water area.According to analyzing and comparing the simulations results among the five algorithms,the hybrid intelligent algorithm by mixing PSO and GA is selected to research the ship path planning under dynamic environment.By bring some dynamic objectives with speed and direction to the static environment model,the dynamic environment model is built in multi-bridges water area.The ship and target ship DCPA(distance to closest point of approach)and TCPA(time to closest point of approach)formulas are proposed through analyzing the ship encounter situations.By setting the minimum safety distance between the ship and dynamic objects,the dynamic collision detection module is designed.Based on ship static path planning,using matlab software,adding to time constraint,the simulations for ship dynamic path planning are carried out under three cases including the ship overtaking the dynamic object,the small angle cross and large angle cross between the ship and the dynamic objects.Analyzing the simulation results and the processes of collision avoidance and overtaking,the validity of the algorithm is verified.