Michael van Lent

AAMAS Conference 2007 Conference Paper

Creating Densely Populated Virtual Environments

• Ryan McAlinden
• Don Dini
• Chirag Merchant
• Michael van Lent

Few virtual environments are capable of supporting large numbers of autonomous agents (> 5000) with complex decision-making on a single machine. This demonstration depicts such an agent infrastructure set within a game-based virtual environment. The embodied agent framework consists of two primary components: a lower-level navigation layer consisting of commercially-available AI middleware, and a higher-level cellular automata system driven by agent goals, resources and thresholds. The overarching gamebased infrastructure consists of these two AI components, along with an ICT-developed perception system sitting atop the Gamebryo rendering engine. The typical number of agents supported on a dual-core CPU with a modern graphics card is ~10, 000 rendering at 30 frames-per-second. To support this quantity and level of intelligence several design considerations were implemented, including the use of multiple threads, a clone/sprite-based avatar view, and a dynamic level-of-detail update system. Future work includes distributing the AI mechanism across multiple machines to support numbers of agents a level of magnitude higher than is currently possible.

AAAI Conference 1999 Conference Paper

Intelligent Agents in Computer Games

• Michael van Lent
• John Laird
• Josh Buckman
• Joe Hartford
• Steve Houchard
• Kurt Steinkraus
• Russ Tedrake
• University of Michigan

As computer games become more complex and consumers demand more sophisticated computer controlled opponents, game developers are required to place a greater emphasis on the artificial intelligence aspects of their games. Our experience developing intelligent air combat agents for DARPA has suggested a number of areas of AI research that are applicable to computer games. Research in areas such as intelligent agent architectures, knowledge representation, goal-directed behavior and knowledge reusability are all directly relevant to improving the intelligent agents in computer games. The Soar/Games project at the University of Michigan Artificial Intelligence Lab has developed an interface between Soar (Laird, Newell, and Rosenbloom 1987) and the commercial computer games Quake II and Descent 3. Techniques from each of the research areas mentioned above have been used in developing intelligent opponents in these two games.

ICML Conference 1999 Conference Paper

Learning Hierarchical Performance Knowledge by Observation

• Michael van Lent
• John E. Laird

AAAI Conference 1996 Short Paper

Constructive Induction of Features for Planning

• Michael van Lent

Constructive induction techniques use constructors to combine existing features into new features. Usually the goal is to improve the accuracy and/or efficiency of classification. An alternate use of new features is to create representations which allow planning in more efficient state spaces. An inefficient state space may be too fine grained, requiring deep search for plans with many steps, may be too fragmented, requiring separate plans for similar cases, or may be unfocused, resulting in poorly directed search. Modifying the representation with constructive induction can improve the state space and overcome these inefficiencies. Additionally, since most learning systems depend on good domain features, constructive induction will compliment the action of other algorithms.