About the IPC

The aim of the Intention Progression Competition is to incentivise and facilitate research around the multi-faceted problem of intention progression.

The competition will:

  1. provide common terminology, data formats, and problem instances;
  2. allow comparison of approaches to the IPP in a scientific manner;
  3. foster software engineering and robustness of approaches; and
  4. raise the academic and public profile of the IPP.

Who should enter

Researchers working in multi-agent systems and the AI planning and scheduling communities, particularly those working on intention progression, and real-time planning and scheduling.

How to enter

Competition entries take the form of a solver for intention progression problems. The solver forms a part of a simple agent that operates in a simulated environment. The agent, environment and problem instances are specified as part of the competition.

At runtime, the agent is asked to achieve a set of goals using a set of pre-specified plans (given in the form of goal-plan trees) in the environment. Not all goals in a problem instance are given initially: the agent may be asked to achieve additional top-level goals at run time.

The agent in turn asks the solver which intention to progress next. The solver must return the next basic action in one of the agent's currently active plans (if execution has reached a subgoal in a plan, this involves choosing an appropriate plan for the subgoal and returning the first action of that plan). The agent and solver communicate via a socket using a simple JSON/XML API.

The agent then returns the selected action to the environment for execution. If the preconditions of the action hold in the current environment state, the environment is updated with the action's postconditions. The environment is dynamic: in addition to applying the postconditions of the action, the environment may change spontaneously.

The cycle repeats until all the agent's goals have been achieved, none of the agent's intentions can be progressed or a timeout is reached.

How to find out more

Check out the Downloads area for information about data formats, sample problems, and submission requirements. Answers to some common questions can be found in the FAQ.

IPC Rules

If you have questions please contact us.

Registering for the competition

To register your interest in the IPC, complete the form on the registration page. You can enter as a team, however only one member of the team should register on behalf of the team.

Following registration, you will receive access to the entrants' portal which contains all of the relevant competition resources. You will also be assigned a Virtual Machine for the purposes of testing your entry against the competition simulator.

How the IPC is organised

The first IPC is being organised by a group of volunteers interested in the Intention Progression Problem. The organising committee is responsible for preparing intention progression problems used in the competition, hosting the competition platform, evaluating the final submissions, and making the results and submissions available after the competition.

Simon Castle-Green University of Nottingham
Alexander Dewfall University of Nottingham
Brian Logan University of Nottingham

The IPC activity is overseen by an international steering committee. The role of the steering committee is to agree on the rules of the competition and recruit an organising committee for each competition. Members of the steering committee have fixed terms, and are periodically replaced by organisers and/or entrants who have participated in several editions of the competition.

Steering Committee
Rafael Bordini PUCRS, Brazil
Mehdi Dastani Utrecht University, Netherlands
Nick Hawes University of Oxford, UK
Gal Kaminka Bar Ilan University, Israel
Brian Logan University of Nottingham
Miquel Ramirez Javega University of Melbourne, Australia
Alessandro Ricci University of Bologna, Italy
Mak Roberts Naval Research Laboratory, USA
John Thangarajah RMIT University
Michael Winikoff Victoria University of Wellington, New Zealand
Neil Yorke-Smith Delft University of Technology