DirectIA

Human behavior modeling is the new challenge simulation technologies have to address. DirectIA is a decision engine based on innovative Artificial Intelligence (AI), used for modeling autonomous entities that exhibit adaptive behaviors. Such autonomous entities are used today in a variety of simulations. They open the way to a new generation of applications, in which human behavior is truly and accurately represented.
Adding the human dimension to simulation
Human behavior modeling is the new challenge simulation technologies have to address. The realism of virtual environments has tremendously improved in the past years, due to the emergence of new technologies and the dramatic increase in computing power available at a small cost. However, most human behavior simulations still rely on algorithms, such as finite state machines or decision trees. These legacy Artificial Intelligence technologies face intractable issues, such as combinatorial explosion, when confronted with real world modeling problems.
While users’ expectations rise, new technologies must be developed and validated to enrich virtual environments with realistic human behavior. This human dimension must be integrated both at the individual- and group-level. For example, a military command post training simulation will require the modeling of the chain of command, while an urban virtual environment will require the simulation of individual or crowd behaviors.
Insert Picture Here.
While users’ expectations rise, new technologies must be developed and validated to enrich virtual environments with realistic human behavior. This human dimension must be integrated both at the individual- and group-level. For example, a military command post training simulation will require the modeling of the chain of command, while an urban virtual environment will require the simulation of individual or crowd behaviors.
Is it possible to use the same technological approach to address such different problems? The goal of this paper is to introduce a new approach for human decision modeling that has been developed using an innovative AI technology. This new approach offers the ability to simulate software entities which display realistic human behaviors. We will show how this technology has been developed to become DirectIA, a new decision engine based on a modular architecture, which can be configured using a standard scripting language, Lua.
The two ways to AI
Classically, a software entity is defined as a simulated element, able to act on itself and on its environment, and which has an internal representation of itself and of the outside world. An entity can communicate with other entities, and its behavior is the consequence of its perceptions, its representations, and its interactions with the other entities.
Simulating entities in a virtual environment requires simulating the entire process that goes from a perception of the environment, or more generally from a stimulus, to an action on the environment. This process is called the AI loop (see figure), and technology used to simulate it can be subdivided in two categories. Sensorimotor or low-level AI deals with either the perception problem (what is perceived?) or the animation problem (how actions are executed?). Decisional or high-level AI deals with the action selection problem (what is the most appropriate action in response to a given perception, i.e. what is the most appropriate behavior?).
While most of the AI technologies available on the market focus on sensorimotor AI, the DirectIA engine is a decisional AI technology which is aimed at solving the problem of action selection in complex synthetic environments.
Traditional or symbolic AI
There are two main approaches in decisional AI. The vast majority of the technologies available on the market, such as planning algorithms, finite state machines (FSA), or expert systems, are based on the traditional or symbolic AI approach. Its main characteristics are:
* It is top-down: it subdivides, in a recursive manner, a given problem into a series of sub-problems that are supposedly easier to solve.
* It is knowledge-based: it relies on a symbolic description of the world, such as a set of rules.
However, the limits of traditional AI, which goal is to build systems that mimic human intelligence, are well-known: inevitably, a combinatorial explosion of the number of rules occurs due to the complexity of the environment. In fact, it is impossible to predict all the situations that will be encountered by an autonomous entity.
Situated or behavioral AI
In order to address these issues, another approach to decisional AI, also known as situated or behavioral AI, has been proposed. It does not attempt to model systems that produce deductive reasoning processes, but rather systems that behave realistically in their environment. The main characteristics of this approach are the following:
* It is bottom-up: it relies on elementary behaviors, which can be combined to implement more complex behaviors.
* It is behavior-based: it does not rely on a symbolic description of the environment, but rather on a model of the interactions of the entities with their environment.
The goal of situated AI is to model entities that are autonomous in their environment. This is achieved thanks to both the intrinsic robustness of the control architecture, and its adaptation capabilities to unforeseen situations.
The situated AI community has presented several solutions to the action selection mechanism. The first attempt to solve this problem goes back to subsumption architectures, which were in fact more an implementation technique than an algorithm. However, this attempt paved the way to several others, in particular the free-flow hierarchies and activation networks. A comparison of the structure and performances of these two mechanisms demonstrated the advantage of using free-flow hierarchies in solving the action selection problem. However, motor schemas and process description languages are two other approaches that have been used with success for autonomous robots.
The DirectIA decision engine
The DirectIA decision engine takes advantage of the situated AI approach. This engine, based on a patented technology allows simulating autonomous entities in their environment. It is a two-layer architecture: a decision layer propagates decisional information to an action layer, which outputs the most adapted action according to the current situation. Thus, this architecture implements an action selection mechanism.
The advantages of this action selection mechanism are manifold:
* Several action possibilities can be explored in parallel before action selection: there is no decisional information loss.
* Several tasks can be handled simultaneously: the system can exhibit compromise behaviors.
* Decision rules in nodes are simpler since they are responsible for decisional information modulation and not sub-choice selection.
* Complex behaviors emerge from the interactions of a small set of decision rules.
As a result, a DirectIA-enhanced entity will choose its actions according to its internal state, its perception of the environment, its memory of past events and locations, and information obtained from communicating with other entities.
Notes and References

Main Menu

See Also

Main Menu