computational robotics

from tasks to motions

Adaptive Mission and Motion Planning to Enhance Autonomy of Underwater Vehicles

(collaboration with the US Naval Research Laboratory)

This work seeks to enhance the autonomy of underwater vehicles both in terms of mission and motion-planning capabilities. By quickly adapting mission and motion plans in response to unanticipated obstacles, changes in the environment, or new information gathered during exploration, the proposed approach seeks to provide the AUV with decision-making mechanisms that will greatly enhance its capabilities to carry out sophisticated missions.

The approach automatically plans low-cost, collision-free, and dynamically-feasible motions that enable an AUV to carry out missions expressed as formulas in temporal logic. The key aspect of the proposed approach is its use of roadmap abstractions in configuration space to guide the expansion of a tree of feasible motions in the state space. This makes it possible to effectively deal with challenges imposed by the vehicle dynamics and the need to operate in the littoral zone, which is characterized by confined waterways, shallow water, complex ocean floor topography, varying currents, and miscellaneous obstacles. Experiments with accurate AUV models carrying out different missions show considerable improvements over related work in reducing both the running time and solution costs.

Related Publications

  • Plaku E and McMahon J (2014): “Motion Planning and Decision Making for Underwater Vehicles Operating in Constrained Environments in the Littoral.” Springer LNCS Towards Autonomous Robotic Systems, vol. 8069, pp. 328–339  [publisher]  [preprint]
  • McMahon J and Plaku E (2014): “Combined Task and Motion Planning for AUVs.” Workshop on AI and Robotics, IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 17–18  [publisher]  [preprint]
  • McMahon J and Plaku E (2013): “Combined Mission and Motion Planning to Enhance Autonomy of Underwater Vehicles Operating in the Littoral Zone.” Workshop on Combining Task and Motion Planning, IEEE International Conference on Robotics and Automation, pp. 17–22  [publisher]  [preprint]
  • McMahon J and Plaku E (2013): “Motion Planning with Linear Temporal Logic for Underwater Vehicles Operating in Constrained Environments.” Workshop on Planning in Continuous Domains, International Conference on Automated Planning and Scheduling, pp. 3  [publisher]  [preprint]