Combining epistemic planning and belief-based programming for collaborative task planning in stochastic domains

To combine state-of-the-art epistemic planning with belief-based stochastic programming techniques for use in collaborative task planning scenarios
Description of the Project: 

This project will combine advances from epistemic planning (embodied in planners such as PKS) with high-level programming constructs enabling the representation of beliefs and stochastic information (such as the ALLEGRO language) for the purpose of collaborative task planning. A new planning system will be built, supporting the generation of multiagent and human-in-the-loop planning in dynamic real-world environments, such as those involving collaborative human-robot interaction to build plans for shared goals, where each agent (artificial or human) carries out part of the planned activities based on their abilities. Appropriate interfaces to these tools will be developed in ROS to integrate them with existing robot simulators and ecosystems.

Resources required: 
High-specification computing environment for planning in simulated environments
Project number: 
First Supervisor: 
Heriot-Watt University
Second Supervisor(s): 
First supervisor university: 
Heriot-Watt University
Essential skills and knowledge: 
Strong programming skills
Desirable skills and knowledge: 
Knowledge of recent planning systems, and belief-based programming, familiarity with ROS

R. Petrick and M.E. Foster (2013). Planning for Social Interaction in a Robot Bartender Domain, ICAPS.

V. Belle and H.J. Levesque (2015). ALLEGRO: Belief-Based Programming in Stochastic Dynamical Domains, IJCAI.

C. Muise, V. Belle, P. Felli, S. McIlraith, T. Miller, A. Pearce, and L. Sonenberg (2015). Planning Over Multi-Agent Epistemic States: A Classical Planning Approach, AAAI.

T. Bolander (2017). A Gentle Introduction to Epistemic Planning: The DEL Approach, arXiv:1703.02192.