Ships Track Keeping Control Based On Quantum Neural Network

Quantum Neural Network is a novel theory by the combination of artificial neural network and quantum computation theory.It can make full use of the advantages of quantum computation,such as the superposition of qubits,quantum parallel computation,and can be used to overcome some inherent shortages of traditional artificial neural network.It can be inferred that,the quantum neural network will be a very important intelligent control schema in the future.Autopilot is one of the most important ship control equipment.The performance of the ship autopilot might affect the stability and safety of the ship navigation.With the rapid development of the shipping industry,science and technology,people have higher and higher request for the ship autopilot.Track keeping control is one of the main functions of the ship autopilot function.Hence,the better performance of the ship track keeping controller is required to continuous exploration.On the basis of the traditional artificial neural network model,a quantum neural network model originated from traditional BP neural network is proposed in this paper.Firstly,a PID course keeping control system with good control performance is designed.Then take this course keeping PID controller as a teacher controller,carry on BP neural network controller and quantum BP neural network controller training work.Furthermore,the indirect track keeping control strategy will be applied for the ship tracking problem in the following part.The LOS(Line of Sight)guidance method will be used to calculate the desired course angle for ship track keeping.And the author applies the ahead turning distance which is based on the maneuverability index K,T and speed loss during the ship turning.In order to overcome the problem that nonlinear PID controller is not effective when ship is at a large turning angle.The author proposes a turning strategy,supposing three virtual turning points to guide the ship to complete the track keeping control task when ship is at a large turning angle,then get a better track keeping controller with better control performance.Then the controller is used as the teacher controller,and the quantum BP neural network controller is trained.Finally,Simulation results show that the convergence and robustness of the quantum neural network track keeping controller is better than the traditional BP neural network controller.The track keeping controller with improved turning strategy has better track keeping effect.