Rym Z. Wenkstern

DIVAs 4.0: A Multi-Agent Based Simulation Framework

In this paper we present DIVAs 4.0, a framework that supports the development of large-scale agent-based simulation systems where agents are situated in open environments. DIVAs includes high-level abstractions for the definition of agents and open environments, a micro kernel for the management of the simulation workflow, domain-specific libraries for the rapid development of simulations, and reusable, extendable components for the control and visualization of simulations. We illustrate the use of DIVAs through the development of a simple simulator where virtual agents are situated in a virtual city.

Virtual Agent Perception in Multi-agent Based Simulation Systems

In this paper we discuss a modular multi-sense perception system for DIVAs virtual agents. It is based on the idea that local environmental in???uences are constantly sensed by one or more of an agent’s senses (sight, hearing, smell, etc.). The perception system is extensible and modifiable and allows users to dynamically modify how an agent perceives its environment at execution time. This paper focuses on the modularity of the agent perception architecture rather than the perception algorithms themselves.

Agent vision in multi-agent based simulation systems

In this paper, we propose virtual agent vision perception techniques to approximate realistic vision while maintaining low execution time. We discuss virtual agent vision in large scale multi-agent based simulations where agents are situated in open environments (i.e., inaccessible, non-deterministic, dynamic, continuous). When dealing with open environments, the efficiency of agent vision algorithms is of great importance since every agent’s perception must be calculated in simulated real-time. We discuss optimizations for vision algorithms in DIVAs, a large scale multi-agent based simulation framework.

Context-Aware Virtual Agents in Open Environments

This paper presents a model for the interaction between context-aware virtual agents and the environment in which they are situated. This model applies to multiagent based simulation systems dealing with human-like virtual agents in decentralized, continuous, and dynamic environments. The model supports an extensible agent perception module, allowing agents to perceive their environment through multiple senses (sight, hearing, smell, etc.). The environment reacts to agent influences as well as user-invoked stimuli by combining these influences to determine the next state of the environment. This paper introduces a formalization and an implementation of the model and discusses multiple scenarios involving context-aware virtual agents situated in dynamic environments.

Simulation of Traffic Network Re-Organization Operations

In this paper we present a model for an Agent-based intelligent Transportation System (ATS). In ATS, the traffic environment is partitioned into areas controlled by specialized agents called service managers. These agents are supervised by a traffic manager agent whose responsibility is to identifying global traffic management strategies. Intersections and vehicles are also equipped with agents. ATS is based on the premises that traffic management is the result of interactions between the various types of agents. We give an overview of MATISSE 2.0, a large-scale multi-agent based traffic simulation platform developed to test traffic scenarios in ATS. We discuss the underlying structure of the traffic network in MATISSE and provide a detailed discussion on the atomic network re-organization operations, i.e., incoming traffic flow elimination, road reversal and crossing elimination. Unlike many existing simulators which execute pre-defined network strategies, MATISSE allows ATS agents to select and execute network re-organization operations dynamically at run-time. The simulation of an evacuation scenario in a traffic network consisting of 384 roads and 64 intersections shows that the traffic re-organization operations improve evacuation time but are not be optimal.

Agent-based decentralized traffic signal timing

In this paper we discuss a multi-agent cooperative approach for decentralized, coordinated traffic systems. Intersection controllers are equipped with agents, i.e., autonomous software systems which are capable of communicating and cooperating with one another to achieve an individual or global goal. Our approach is based on real-world traffic parameters and constraints, and is meant to be implemented in existing traffic systems with minimal changes. Experimental results show that the agent-based approach outperforms the traditional pre-timed and actuated modes when traffic is heavy.

On Modeling and Verification of Agent-Based Traffic Simulation Properties in Alloy

The advances in Intelligent Transportation Systems ITS call for a new generation of traffic simulation models that support connectivity and collaboration among simulated vehicles and traffic infrastructure. In this paper we introduce MATISSE, a complex, large scale agent-based framework for the modeling and simulation of ITS and discuss how Alloy, a modeling language based on set theory and first order logic, was used to specify, verify, and analyze MATISSEs traffic models.