« Older Home
Loading Newer »

Archive for the 'ai' Category

Interactive Objects in Gaming Applications (Sapienza University of Rome-2013)

I am teaching a topics course with Prof. Marco Fratarcangeli about “Interactive Objects in Gaming Applications: Basic principles and practical scenarios using the Unity platform”, in which I am covering the part about how objects can exhibit behavior. In particular I focus on i) Pathfinding ii) Action-based decision making, and iii) AI Architectures.

i) Pathfinding


The lectures about path planning (or pathfinding) can be found on slideshare: Pathfinding-Part1, Pathfinding-Part2, Pathfinding-Part3. Also, the Unity code examples of Part 2 (based on Aron’s Pathfinding and Unity’s 3D Platformer Tutorial) can be found here: Pathfinding-Unity.zip

Information about the material will be updated as we go, as well as a long list of references.

Abstract of the course:

The course aims at providing a hands-on introduction on the basic principles as well as state-of-the-art methods and techniques for building interactive objects in video games. Each of the aspects examined will be presented both in a theoretical framework that allows studying related research problems as well as in a practical setting providing the tools needed for a real-world implementation. For the latter we will rely mostly on Unity, a popular game engine that is becoming an emerging standard for indie game development. The course is divided in two main themes: one that focuses on the physical aspect of the behavior of interactive objects, and one that focuses on the artificial intelligence aspect of the behavior of smart objects.

In the first part we will introduce the basic concepts for the development of interactive applications, including logic, rendering, physics, audio, user interaction and graphical interface. We will map each of these concepts in their practical implementation by using the Unity toolset.

In the second part we will continue with some of the most widely used methodologies that allow smart objects and non-player characters (NPCs) to exhibit autonomy and flexible behavior through various forms of decision making, including techniques for pathfinding, reactive behavior through automata and processes, and goal-oriented action planning.

Video-games and more in the upcoming “Planning in Games ICAPS-2013 Workshop” (PG-2013)

Interested in applications of AI planning techniques to video-games and more? Take a look at the upcoming Planning in Games Workshop at ICAPS-2013 to take place in Rome!

The workshop aims at gathering researchers and practitioners interested in the use of planning in games to discuss current work and future directions: from classic games to video games, from academia to the industry, from researchers to developers and designers to gamers, from path planning to strategic planning, and more.


Call for papers

Topics and Objectives

The Planning in Games Workshop covers a wide range of topics related to developing, integrating, and benchmarking single agent, adversarial, and multi-agent planning techniques into classic games as well as video games. While the emphasis of the workshop is on the methods and techniques developed in the field of AI, the workshop aims at bringing together researchers and practitioners involved in applications of planning in the field of video games, to discuss current work and future directions. We especially welcome discussions and demonstrations of existing systems.

Artificial Intelligence Planning is successfully used in video-games: heuristic-based STRIPS-like planning and HTN Planning generate character behavior in several fast paced games since 2005, reaching millions of players. This certainly does not make planning in games a solved problem: from new game genres to next-generation consoles and new markets such as cloud gaming, the AI Planning research frontier is wide and open to any kind of planning technique in a gaming context.

This 3rd edition of the ICAPS Workshop on Planning in Games shall acknowledge the tighter link with the video-game industry, while aiming at inspiring traditional Game AI Planning research such as optimal planning in huge search spaces and temporal reasoning, and traditional games such as go and chess: this workshop invites submissions on any aspect of AI planning in the largest possible domain of games.

Relevant topics include:

  • Strategic and tactical planning
  • Planning techniques for abstract games such as board games,
  • Planning for command hierarchies
  • Temporal and spatial planning
  • Real-time planning and replanning
  • Drama and story planning
  • Checking and debugging game design with planning
  • Cloud-based planning services
  • Plan recognition, cased-based and crowd-sourcing planning
  • Abstractions and representational issues
  • Competitions

Like previous Planning in Games workshops, we hope PG-2013 to be a highly participatory and discussion-oriented forum. We anticipate a one-day workshop that will comprise of several sessions including presentations of research papers, position papers, and posters, and a panel discussion.

Invited Talks

  • Dana S. Nau (University of Maryland)
  • Alex Champandard (AIGameDev.com)
  • William van der Sterren (CGF-AI)


Potential participants are invited to submit either a full
length technical paper or a statement of interest with a position
paper. Submissions are accepted in PDF format only, using the AAAI
formatting guidelines at:

Submissions must be no longer than seven (7) pages in length,
including references and figures. Author names should be included.
Overlength papers will not be accepted for publication. Papers must
be submitted electronically in PDF format by the due date.

Submission website:

Electronic proceedings will be offered to delegates in a USB key
by ICAPS-2013, and will also be published online in the ICAPS-2013


Submission deadline: March 23, 2013
Notification: April 20, 2013
Final version: May 10, 2013
Workshop date: June 11, 2013 (TBC)

Workshop Program Chairs

Program Committee

  • Marc Cavazza, University of Teesside, United Kingdom
  • Carle Cote, Eidos, Canada
  • Luke Dicken, University of Strathclyde, United Kingdom
  • Alan Fern, Oregon State University, United States
  • Peter Gregory, University of Teesside, United Kingdom
  • Carlos Linares López, Universidad Carlos III de Madrid, Spain
  • Christian Muise, University of Toronto, Canada
  • Héctor Muñoz Ávila, Lehigh University, USA
  • Jeff Orkin, MIT Media Lab, United States
  • Julie Porteous, University of Teesside, United Kingdom
  • Mark Riedl, Georgia Institute of Technology, USA
  • William van der Sterren, CGF-AI, Neatherlands

Report on the 8th International Workshop on Cognitive Robotics (CogRob-2012)

The 8th International Workshop on Cognitive Robotics was held at Toronto, Ontario, Canada at the Sheraton Centre Toronto, on July 22–23, 2012. Cognitive robotics is concerned with integrating reasoning, perception and action with a uniform theoretical and implementation robotics framework. The goal of this workshop was to bring together researchers involved in all aspects of the theory and implementation of cognitive robots to discuss current work and future directions.

Wolfram Burgard (Institute for Informatik, Albert-Ludwigs-Universitat Freiburg, Freiburg, Germany), Kurt Konolige (Industrial Perception, Inc., Palo Alto, USA), Maurice Pagnucco (University of New South Wales, Sydney, Australia), and Stavros Vassos (National and Kapodistrian University of Athens, Athens, Greece) served as co-chairs for this workshop. The papers of the workshop were published as AAAI Press Technical Report WS-12-06.

The use of both software robots (softbots) and robotic artifacts in everyday life is on the upswing and we are seeing increasingly more examples of their use in society with commercial products around the corner and some already on the market. As interaction with humans increases, so does the demand for sophisticated robotic capabilities associated with deliberation and high-level cognitive functions. Combining results from the traditional robotics discipline with those from AI and cognitive science has and will continue to be central to research in cognitive robotics. As improvements in technology lead to more sophisticated sensors, actuators, etc. together with a decrease in size, power consumption and cost, the need for higher-level cognitive functions on these more sophisticated robots is becoming more and more pressing.

Research in robotics has traditionally emphasized low-level sensing and control tasks including sensory processing, path planning, and manipulator design and control. In contrast, research in cognitive robotics is concerned with endowing robots and software agents with higher level cognitive functions that enable them to reason, act and perceive in changing, incompletely known, and unpredictable environments. Such robots must, for example, be able to reason about goals, actions, when to perceive and what to look for, the cognitive states of other agents, time, collaborative task execution, and so on. In short, cognitive robotics is concerned with integrating reasoning, perception and action within a uniform theoretical and implementation framework.

This two-day workshop brought together researchers from a variety of subfields of AI in addition to researchers in the field of cognitive science. The 15 papers that were presented in the workshop focused on a broad range of topics from theoretical approaches and methodologies for various aspects of cognitive robotics to practical and challenging applications of cognitive robots in the real world.

Some of the topics discussed by the presenters include practical languages for reasoning about action and change, agent programming languages for high-level behavior, plan recognition, modelling visuospatial abilities, as well as the framework of conceptual spaces, the Soar cognitive architecture, and emotional intelligence. As far as practical applications of cognitive robotics is concerned, the discussions included an application in personalized guided tour using cognitive robots (“Personalized Guided Tour by Multiple Robots through Semantic Profile Definition and Dynamic Redistribution of Participants” by Anna Hristoskova, Carlos Aguero, Manuela Veloso and Filip De Turck), a framework for modelling the skills of manufacturing robots (“A Taxonomic Framework for Task Modeling and Knowledge Transfer in Manufacturing Robotics” by Jacob Huckaby and Henrik I. Christensen), as well as a prototype robotic bartender (“What Would You Like to Drink? Recognising and Planning with Social States in a Robot Bartender Domain” by Ron Petrick and Mary Ellen Foster).

In addition to the contributed presentations, the workshop included two invited talks from renowned members of the AI community that took place at the beginning of each day of the workshop. Michael Thielscher (University of New South Wales, Australia) talked about the research topic of General Game Playing in the context of cognitive robotics, and Peter Stone (University of Texas, USA) focused on the RoboCup competition discussing its evolution in the recent years and the major challenges that have been identified.

The discussions that were initiated by the invited talks and the paper presentations also continued in a panel session that took place near the end of the second day of the workshop. The panel included four panelists, Giuseppe De Giacomo (Sapienza University of Rome, Italy), Unmesh Kurup (Carnegie Mellon University, USA), Gerhard Lakemeyer (Aachen University of Technology, Germany), and Peter Stone (University of Texas, USA), and the topic of the discussion was “Practical challenges for Cognitive Robotics”.

Three major themes emerged from a lively discussion among the panelists and the participants of the workshop identifying the following as major challenges: i) the ability to form ad-hoc teams of cognitive robots for collaboratively addressing a given situation, ii) the use of software worlds, such as video-games, not only as a testbed for experimenting and evaluating approaches for cognitive robotics, but also as a self-standing application domain that becomes ubiquitous is the modern age of information, and iii) an effective account of perception that incorporates both symbolic and non-symbolic reasoning also taking into account a fusion of the available vast amount of information in existing knowledge bases.


Action-Based Imperative Programming with YAGI (CogRob-2012)

Action-Based Imperative Programming with YAGI, Alexander Ferrein, Gerald Steinbauer, Stavros Vassos, In Proceedings of the 8th International Workshop on Cognitive Robotics (CogRob-2012) of the 26th AAAI Conference (AAAI-2012) Conference, Toronto ON, Canada, 2012.
[pdf | citeulike| slides | slideshare]

Continue reading ‘Action-Based Imperative Programming with YAGI (CogRob-2012)’

Introduction to STRIPS Planning and Applications in Video-games (SapienzaUniversity-2012)

In the last few weeks I have been in Italy, teaching a short course related to Artificial Intelligence and Video-games.

The course  was offered in the Ph.D. program in Computing Science and Engineering of the Sapienza University in Rome, Department of Computer, Control, and Management Engineering.  Part of the material covered was also presented as invited lectures in the Artificial Intelligence in Games and Videogames M.Sc. course taught by Prof. Marco Schaerf.

The course aims to provide an introduction to the techniques currently used for the decision making of non-player characters (NPCs) in commercial video games, and show how a simple deliberation technique from academic artificial intelligence research can be employed to advance the state-of-the art.

The course is fairly introductory mostly aimed for first-year Ph.D. students and M.Sc. students, and is also appropriate for last-year undergraduate students.

All the material presented in lectures is available at tinyurl.com/NPC-AI-LaSapienza.

Real-time Action Planing with Preconditions and Effects (GameCoderMag-2012)

I wrote an article for the March issue of the Game Coder Magazine:

Abstract: In this article I will be covering an artificial intelligence (AI) technique for decision making that can be used in various parts of game development to account for “thinking before acting”. The technique is called Classical Planning in academic AI research, and is one of the most basic approaches for deliberating about the effects of actions and the way the properties of a given domain change under these effects. Variants of this technique have been used successfully in game development under the name Goal Oriented Action Planning (GOAP), each time focusing on a different aspect of this technique, adopting it for the particular needs of the game.

You can get the article here.

More information: In the article I go over a simple example inspired from real-time strategy games. I focus on the behavior of a peasant that can be instructed to bring food or handle other resources, and extend the character’s functionality so that they can take more advanced commands which may need a series of actions to be realized. Using this example I go over one of the simplest forms of planning, that is propositional STRIPS planning, and present an implementation using Python.

The following screenshots are code listings included in the article, which show how the planning algorithm can help the peasant find ways to realize the given commands. 

The self-contained Python code can be found here: PyPlan. The file goap.py contains all the necessary classes and methods for implementing planning, and example.py includes a simple example based on the peasant scenario that is explained in detail in the article.


author = {Vassos, Stavros},
citeulike-article-id = {11193150},
citeulike-linkout-0 = {http://stavros.lostre.org/files/Vassos12ActionPlanning.pdf},
journal = {GameCoder Magazine},
keywords = {goap, planning, strips, video\_games},
month = mar,
number = {3},
pages = {20–27},
title = {Real-time Action Planing with Preconditions and Effects},
url = {http://stavros.lostre.org/files/Vassos12ActionPlanning.pdf},
year = {2012}

NPCs with Artificial Intelligence: From FSMs to BTs to GOAP (GameExpo-2012)

On Friday March 16, I gave a talk at the IEEE Patras Games Expo 2012, which was organized by the IEEE Computer Society Student Chapter in collaboration with the IEEE Student Branch of University of Patras.

My talk was related to artificial intelligence (AI) techniques for specifying the behavior of non-player characters (NPCs) in video games. As a quick introduction to the main ideas behind the most common techniques that have been used in commercial video games, I went over three approaches: Finite State Machines (FMSs), Behavior Trees (BTs), and Goal-Oriented Action Planning (GOAP). The intention was to introduce students to the ideas and highlight what I consider to be the past, the present, and the future of techniques for NPC behavior.

The slides of my talk (in Greek) can be found  here. Some examples in the slides are taken from the book Artificial Intelligence for Games by  Ian Millington, and John Funge, an excellent resource for all types of AI methods for game development.


Organizing the 8th International Cognitive Robotics Workshop at AAAI-2012, Toronto, Canada

Our proposal for organizing the 8th International Cognitive Robotics Workshop at the upcoming AAAI conference has been accepted! The workshop chairs are Wolfram Burgard from Albert-Ludwigs-Universitat Freiburg in Germany, Kurt Konolige from Willow Garage USA, Maurice Pagnucco from University of New South Wales in Australia and me. Some details follow.

The 8th International Cognitive Robotics Workshop (CogRob-2012) will be held on July 22-23, 2012 in Toronto, Canada, as part of the of the Twenty-Sixth Conference on Artificial Intelligence (AAAI-12) workshop programme.

CogRob is a well-established forum that aims to bring together researchers involved in all aspects of the theory and implementation of cognitive robots, to discuss current work and future directions. While the emphasis of the workshop is on the methods and techniques developed in the field of Artificial Intelligence, we welcome work in related cognitive science disciplines investigating computational/cognitive models of behavior. Also, we especially welcome discussions and demonstrations of implemented systems.

For more information visit the website of CogRob-2012 or the ResearchGate page of CogRob-2012.

Talk about SimpleFPS at SRI International, Menlo Park, CA (2011)

On Friday October 14, I gave a talk about some preliminary work on building a PDDL benchmark for First-Person Shooter games at SRI International’s Artificial Intelligence Center, Menlo Park, CA USA, 2011.

The talk was based on the following workshop paper I presented at AIIDE-2011: The SimpleFPS Planning Domain: A PDDL Benchmark for Proactive NPCs.

The details of the talk can be found here. The slides of my presentation can be found here.

A Database-type Approach for Progressing Action Theories with Bounded Effects (Book chapter-2011)

A Database-type Approach for Progressing Action Theories with Bounded Effects, Stavros Vassos, Sebastian Sardina, In Gerhard Lakemeyer and Sheila McIlraith, editors, Knowing, Reasoning, and Acting: Essays in Honour of Hector J. Levesque, College Publications, 2011.
[pdf | citeulike | more]

Continue reading ‘A Database-type Approach for Progressing Action Theories with Bounded Effects (Book chapter-2011)’