Interleaving Temporal Planning and Execution in Robotics Domains

Book One

Volume

Proceedings of the AAAI Conference on Artificial Intelligence, 19

Robotics

Many autonomous systems such as mobile robots, UAVs or spacecraft, have limited resource capacities and move in dynamic environments. Performing on-board mission planning and execution in such a context requires deliberative capabilities to generate plans achieving mission goals while respecting deadlines and resource constraints, as well as run-time plan adaption mechanisms during execution. In this paper we propose a framework to integrate deliberative planning, plan repair and execution control in a dynamic environment with stringent temporal constraints. It is based on lifted partial order temporal planning techniques which produce flexible plans and allow, under certain conditions discussed in the paper, plan repair interleaved with plan execution. This framework has been implemented using the IxTeT planner and used to control a robotic platform.