Research On Tugboat Path Planning Technology Based On Improved Potential Field Grid Method

More than 90% of the world’s trade and transportation is still completed by sea.With the development of artificial intelligence industry,ship intelligence is the trend of the world today.The advent of smart ships can reduce the workload of crew,improve work efficiency,reduce human errors in operation,and better implement maneuvering and collision avoidance.However,the related technology of smart ship is still in the stage of development and testing,and has not been applied on a large scale.For smart ships,key technologies include awareness,planning,decision-making and control.Among them,route planning is an indispensable and important part of intelligent technology,and its quality directly affects the effect of decision-making and control.In view of this,this thesis studies the automatic obstacle avoidance technology of tug based on electronic nautical charts.Through the comparison and comprehensive analysis of traditional intelligent algorithms,it is found that the artificial potential field method has natural advantages over other algorithms: the planned path is relatively smooth and safe,easy to control in real time,and its disadvantage is easy to fall into local optimal solution.If the traditional artificial potential field method is improved reasonably,a relatively reasonable path will be obtained.Therefore,this thesis will implement the path planning in the process of automatic obstacle avoidance based on the improved artificial potential field method.The main tasks completed are as follows:(1)An electronic nautical chart rasterization method based on nautical chart color matrix is presented.To realize the path planning in the process of tug obstacle avoidance,the electronic nautical charts in the tug operation port area need to be rasterized,and before rasterization,the electronic nautical charts need to be binary.Due to the complexity of the color types of electronic nautical charts,this thesis presents a rasterization method based on the color matrix of nautical charts,that is,extracting the RGB(red,green and blue)color matrix from the nautical charts,and binarizing it into land(black matrix)or navigable waters(white matrix)according to actual needs.(2)A tug obstacle avoidance path planning method based on raster map and improved artificial potential field method is presented.This method introduces a new repulsive potential field function into the existing formula of artificial potential field method and adds a regulatory factor.In order to limit the active area,the boundary repulsion field is introduced.By adjusting the boundary repulsion factor and the adjustment factor,the algorithm can be better prevented from falling into local optimum and the optimization efficiency can be improved.In addition,in order to obtain a continuous and smoother curvature of the path,a cubic spline interpolation is applied to the resulting curve.(3)The method proposed in this thesis is simulated using MATLAB and compared with the traditional algorithm.Two different scenarios,simple and complex,were used to validate the test.The test results show that the improved artificial potential field method has a smaller curvature,a 33.3% reduction in turning points and a 3.1% reduction in voyage distance under simple obstacle environment compared with the traditional artificial potential field method.Under complex environment conditions,traditional artificial potential field method will fall into local optimum,and there will be more oscillations.The improved artificial potential field method has obvious advantages over the traditional algorithm,and the path length is shortened by 3.3%,the turning point is significantly reduced by 61.5%,and the curvature is smaller.The route planning algorithm based on the improved artificial potential field method proposed in this thesis can be used for route planning and real-time obstacle avoidance in the process of intelligent tug operation.