Task Scheduling and Trajectory Generation of Multiple Intelligent Vehicles

The problem of multi-vehicle path planning can be treated as a machine scheduling or a vehicle routing problem with definite temporal, spatial and other constraints. We consider the scenario of path planning of multiple Automated Guided Vehicles (AGVs) and Automated Trucks (ATs) in a Ship-Container Terminal area*. Our main focus is on developing global and local path planning methods for each of these vehicles along with a centralized task scheduler to designate goals to each of the vehicle. The whole problem can be divided in to two folds: Firstly, it is required to produce optimal schedule (planned path lengths), for a fleet of AGVs to certain goal points to pickup/drop off containers. It is treated similar to a truck scheduling problem with special constraints to deal with dynamic and complex environment. Secondly, it involves finding collision free trajectories for each of the vehicle using gradient descent method. We aim to propose an integrated framework for solving the goal assignment and trajectory planning problem minimizing the maximum cost over all vehicle trajectories and avoiding conflicts.