Research On Automatic Container Vehicle Scheduling And Path Planning In Container Terminals

Intellectualization and networking are the current trends in the development of vehicle technology.Scientific research institutes and automotive enterprises have set off a wave of research on vehicle networking and autonomous driving technology.At this stage,the degree of intelligence and networking does not yet support automatic driving under all road conditions.The application of Internet of Vehicles technology and automatic driving technology in closed scenarios such as ports,mines,airports,parks,etc.is currently a consensus in the industry.Large scale repetitive transportation closed scenarios currently face problems such as high energy consumption costs and frequent vehicle conflict congestion.Designing multi-objective scheduling and path planning algorithms is crucial to reducing energy consumption costs and conflict free path planning.This article conducts research on the minimum energy consumption problem and conflict free path planning problem of automatic container vehicle transportation in the terminal scenario,while considering the scenario of sudden changes in tasks.The specific research content is as follows:(1)Based on the vehicle transportation scenarios of the world’s mainstream automated container terminals,the transportation scenario in this paper is designed and a multi constraint model is established,laying the foundation for the implementation of vehicle scheduling and path planning algorithms.Firstly,the layout of the wharf scene in this article is determined and relevant assumptions are made.Secondly,a random generation mechanism for container transportation tasks is set up.Finally,on the basis of decomposing and simplifying the task transportation process,a multi constraint model is established with the goal of completing the terminal transportation task with minimum energy consumption.(2)In order to reduce the cost of vehicle energy consumption,a simulated annealing genetic algorithm based on combination strategy was designed.First,the shortest path between different loading and unloading nodes is calculated in advance through the Freudian algorithm,and the resulting distance matrix can be directly invoked during subsequent scheduling algorithm calculations,improving time efficiency.Secondly,a genetic algorithm based on combination strategy is designed,including task combination,population initialization,selection,crossover,mutation,and other links.After that,multiple sets of container transportation tasks were randomly generated,and two scenarios,task combination and task non combination,and vehicle first completed transportation and genetic algorithm,were compared.The energy consumption of autonomous guided vehicle transportation under four scenarios was obtained and the four scenarios were compared and analyzed.Finally,simulated annealing algorithm is used to optimize the genetic algorithm,and the effectiveness of simulated annealing genetic algorithm based on combination strategy is verified.(3)A set of automatic guided vehicle conflict resolution scheme for pre task execution and task execution process is designed.Firstly,based on the usage of automatic guided vehicles,four conflict scenarios are analyzed: frontal conflict,cross conflict,occupancy conflict,and chasing conflict.Secondly,a vehicle conflict processing process before task execution is designed,using the time window principle to detect conflicts before task execution,and four conflict resolution schemes are designed,including setting the road traffic direction,selecting other candidate paths,slowing down,parking,and waiting,and task rescheduling planning.After that,the vehicle collision constraint model,safety distance,speed,and location are used to detect conflicts during task execution,and conflicts are resolved through deceleration and braking,intersection priority judgment,and other methods.Finally,using open TCS simulation software to verify the vehicle conflict resolution before task execution,and using Prescan software to verify the safe distance during task execution.(4)Three task rescheduling strategies are developed and compared for scenarios where the number of tasks changes.Firstly,the scenarios where the number of tasks may change are summarized.Secondly,three task rescheduling strategies are formulated for task change scenarios,including no rescheduling strategy,complete rescheduling strategy,and insert rescheduling strategy.Finally,three task rescheduling strategies are compared and analyzed from four aspects: the time spent executing the algorithm,the energy consumption of the carrier,the impact of the old task sequence,and the applicable scenarios.