Research On AGV Dynamic Path Planning Based On Real Time Obstacle Avoidance

With the advent of information,intelligence and digital manufacturing,enterprises have higher and higher requirements for the automation of material distribution.The efficient and stable operation of material distribution is an effective measure to improve production efficiency,reduce production costs and improve enterprise profits.How to combine vehicle path planning with autonomous obstacle avoidance to achieve dynamic path planning to improve the effectiveness of material distribution scheme is the key direction of enterprise research in material distribution scheme.Material demand time window,material distribution AGV scheduling and cost minimization are not only three important aspects of material distribution design management,but also important means to improve plant production efficiency and equipment utilization.It is an effective way to optimize the material distribution plan by combining the data in the existing material distribution plan with scientific methods to evaluate the distribution time,AGV trolley travel distance and distribution equipment rate of the distribution plan,and then formulating a scientific distribution plan.The research of this paper mainly includes the following contents:(1)Based on the distribution information of material distribution in modern chemical plants,the VPR problem of multi AGV material distribution with time windows is proposed and the VRPTW mathematical model is established by analyzing the quantity of material distribution,the constraints of time windows,and the law of material station consumption.The multiple algorithms of path optimization are analyzed and a hybrid genetic particle swarm optimization algorithm is designed to determine the optimal scheme of material distribution.On this basis,various algorithms and strategies of path optimization are analyzed,and a hybrid optimization algorithm is designed to solve the model.By comparing with the solution scheme of simulated annealing algorithm and the original scheme,the optimization scheme of workshop material distribution is determined.(2)The problem of AGV real-time obstacle avoidance in complex dynamic environment is studied.First,the planning algorithm of local dynamic path is studied,focusing on the realization process of the dynamic obstacle avoidance function of the traditional DWA algorithm,A~* and DWA algorithm are selected to study the dynamic obstacle avoidance of AGV.Secondly,the A~* algorithm is improved to discard harmful nodes and turning points to improve the efficiency of planning.Finally,three examples with increasing complexity are used to simulate the path planning of the algorithm before and after improvement,the effectiveness of the improved algorithm is proved by analyzing its planning time,moving speed,angular speed and other indicators.(3)Aimed at the problem of virtual simulation of multi AGV material distribution,through the analysis of a variety of material distribution simulation software,Flexsim is selected to simulate the material distribution in the electric tool assembly workshop.First,a model of multiple AGV material distribution with time windows is constructed under the original scheme,and its moving time,loading quantity,moving distance,etc.are summarized and output.Secondly,the material distribution scheme based on hybrid genetic particle swarm optimization algorithm is simulated,and its related evaluation indexes are analyzed.Finally,the final optimization scheme is selected by comparing and analyzing the evaluation indicators of the material distribution schemes of the two schemes.