Path Planning for Intelligent autonomous vehicles in container terminals based on dynamic traffic flow assignment

Intelligent autonomous vehicles (IAVs) are considered a new kind of horizontal transporter in container terminals since they are superior to automated guided vehicles (AGVs) in performing more flexible and intelligent maneuvers without the layout of magnetic nails. In contrast to finding the shortest path for every single IAV that may not guarantee overall transportation efficiency, this paper considers a large of container-transportation tasks and proposes a novel path planning algorithm for all the scheduled IAVs based on dynamic traffic flow assignment. We first establish a traffic flow model to describe flow evolution in transportation areas including quay and stack sides in container terminals. Then the minimization of total transportation time is considered as the objective function and path planning for all the scheduled IAVs is designed in the model predictive control (MPC) framework. The routes are obtained by establishing the receding-horizon optimization based on predicted traffic states and solving by a resilient back-propagation solution algorithm. The simulation results show that the proposed algorithm can mitigate congestion in transportation areas by redirecting the IAV flow distribution, improve the overall transportation efficiency and effectively handle the system uncertainties and disturbances, i.e., inaccurate information of operation tasks, facilities failure and breakdown.