Research On Port AGV Path Planning Method Based On Dynamic Window Approach Integrated With A* Algorithm

Automated Guided Vehicles(AGV)is an indispensable technical means in modern logistics and transportation industry,and path planning is one of the key links in which AGV system can be widely used.The so-called path planning means that AGV can determine an optimal path from the initial state to the target state in a specific working environment according to some optimal criteria,and avoid collision with other objects in the course of driving.At present in China,the solidified road network is mostly used in port AGV,and the driving route of AGV is single.Faced with more and more complicated transportation tasks,it is necessary to improve the existing path planning methods and improve the AGV transportation efficiency.Firstly,the dynamic window approach is proposed to solve the static driving problem of port AGV transportation.The simulation results show that dynamic window approach can make AGV achieve good obstacle avoidance function and improve its static route.However,The simulation results also show that in dynamic environment,dynamic window approach can be used to solve the problems of AGV transportation,but it is possible to reach the target point bypass or even fail to complete the work flow.Then,this thesis analyses the reasons why the dynamic window approach appears the above phenomena in the dynamic environment.The reason is that AGV does not consider the global optimality,which results in avoiding the optimal path of arrival in the process of avoiding obstacles.In view of this phenomenon,this thesis proposes a combined algorithm which combines dynamic window approach and A*algorithm.By studying the evaluation function indexes of the two algorithms,a new combined evaluation function is established.The new evaluation function enables the dynamic window approach to adopt the heuristic idea of A*algorithm when evaluating the optimal trajectory,taking into account the global optimality of the path.On the platform of MATLAB,the dynamic window approach,combined algorithm and another local obstacle avoidance algorithm,artificial potential field method,are compared and verified.It is found that the combined algorithm improves the success rate of reaching the target point and the smoothness of the path compared with the dynamic window approach.A large number of experimental results show that the combined algorithm also reduces the operation time to a certain extent.Compared with the potential field method,the proposed method has higher smoothness and is more suitable for large mobile carriers.Finally,in order to further improve the transportation efficiency of AGV,two kinds of AGV operation rules are proposed in this thesis.The experimental results show that adding turning area can effectively avoid AGV in-situ turning.Adding driving rules to AGV with the same group of bridge cranes and different groups of bridge cranes can maximize the use of AGV on limited sites.In addition,considering the application environment of port AGV,the navigation method based on magnetic nail positioning is mostly used.In this thesis,an integrated navigation system based on magnetic nail and inertial navigation is presented.