Decision making for autonomous vehicles in the presence of uncertain information

A routing-scheduling problem in a class of manufacturing systems is considered for a group of autonomous vehicles moving in a common working space to perform a given set of tasks. The vehicles subject to unforeseen events, e.g. disruptions, failures, collisions which should be omitted, requiring the adaptation of originally developed schedules. Two approaches based on proactive and reactive scheduling are proposed. In the former, it is assumed that tasks' execution times are not known precisely, and the minimax absolute regret decision making problem is solved. The latter uses motion control decisions for the group of vehicles to current improvements of the schedules. Two versions are investigated referred to as open-loop and closed-loop decision making systems.