Simultaneous planning and scheduling for multi-autonomous vehicles

Abstract

With the increasing applications of autonomous vehicles in dynamic and strictly constrained environments such as automated container terminals, efficient task/resource allocation and motion coordination (i.e., path and speed planning) of multi-autonomous vehicles has become the critical problem and have therefore been recently recognized as the key research issues by both academics and industry. This chapter addresses a generic approach for integration of task allocation, path planning and collision avoidance, which has so far not attracted much attention in the academic literature. A Simultaneous Task Allocation and Motion Coordination (STAMC) approach is presented. Two metaheuristic algorithms, i.e. simulated annealing and ant colony, and an auction algorithm are investigated and applied. The proposed approach is able to solve the scheduling, planning and collision avoidance problems simultaneously; improve the usage of bottleneck areas; handle dynamic traffic conditions and avoid deadlock. Simulation results demonstrated the effectiveness and efficiency of this approach. © Springer-Verlag Berlin Heidelberg 2007.