Generating Various and Consistent Behaviors in Simulations
Philippe Mathieu2009, Advances in Intelligent and Soft Computing
Sign up for access to the world's latest research
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Abstract
In multi-agent based simulations, providing various and consistent behaviors for the agents is an important issue to produce realistic and valid results. However, it is difficult for the simulations users to manage simultaneously these two elements, especially when the exact influence of each behaviorial parameter remains unknown. We propose in this paper a generic model designed to deal with this issue: easily generate various and consistent behaviors for the agents. The behaviors are described using a normative approach, which allows increasing the variety by introducing violations. The generation engine controls the determinism of the creation process, and a mechanism based on unsupervised learning allows managing the behaviors consistency. The model has been applied to traffic simulation with the driving simulation software used at Renault, SCANeR c II, and experimental results are presented to demonstrate its validity.
Related papers
A behavioral multi-agent model for road traffic simulation
Engineering Applications of …, 2008
Multi-agent systems allow the simulation of complex phenomena that cannot easily be described analytically. Multi-agent approaches are often based on coordinating agents whose actions and interactions are related to the emergence of the phenomenon to be simulated. In this article, we focus on road traffic simulation, specifically the design of a road traffic simulation tool able to deal realistically with road junctions. We propose a multi-agent behavioral model based on (i) the opportunistic individual behaviors that describe the norm violation and (ii) the anticipatory individual abilities of simulated drivers that allow critical situations to be detected. Our proposition has been validated for different traffic scenarios. Specifically, we simulated the traffic in a real intersection and then compared the simulated traffic flow with the real flow to highlight the relevance of our approach.
A versatile description framework for modeling behaviors in traffic simulations
Microscopic simulations of road traffic are a typical application domain for Multi-Agent Systems. Indeed, the individual-based approach allows to take into account the diversity of behaviors so as to consider real situations. More recently, geographical databases provide environmental information under open formats, which offers the opportunity to design agent-based traffic simulators which can be continuously informed of changes in traffic conditions. The use of such data, together with the adaptability of MAS, allows the realization of decision support systems that are able to integrate environmental and behavioral modifications in a direct way, and compare various scenarios built from different hypotheses in terms of actors, behaviors, environment and flows. We describe here a modeling approach and a comprehensive process which lead to the development of such a tool.
Non-normative behaviour in multi-agent system: Some experiments in traffic simulation
Proceedings of the IEEE/ …, 2006
Most of the works related to norms and multi-agent systems focus on the design of normative agents systems making the assumption that agents always respect norms. Our aims in this article are (i) to discuss the relevance of this assumption in some specific contexts and to highlight some benefits of designing non-normative behaviour agents, (ii) to expound the methodology followed in a concrete application which consists in traffic simulation at junction. In particular, based on statistical traffic results, we show how non-normative behaviours contribute to improving the realism of simulation.
Dealing with the right of way in advanced traffic simulation. A characterization of drivers' behaviours in a multi-agent approach
2017
With the emergence of ADAS and autonomous vehicle, the need of simulation software to test advanced systems and models is unavoidable. The objective of this work is to characterize, in a multi-agent approach, the traffic vehicles behaviours, and to model them in a versatile traffic simulator, serving as a testing tool for integration in the traffic module of the SCANeR Studio T M simulation software. The algorithm will bring more natural interactions between vehicles in simulation and allow the emergence of new relevant situations for autonomous vehicles, observable in real, such as collision risk situations or even accidents, which have for now to be scripted in scenarios and do not occur naturally.
An Integrative Modelling Language for Agent-Based Simulation of Traffic
IEICE Transactions on Information and Systems, 2016
Traffic is a key aspect of everyday life. Its study, as it happens with other complex phenomena, has found in simulation a basic tool. However, the use of simulations faces important limitations. Building them requires considering different aspects of traffic (e.g. urbanism, car features, and individual drivers) with their specific theories, that must be integrated to provide a coherent model. There is also a variety of simulation platforms with different requirements. Many of these problems demand multidisciplinary teams, where the different backgrounds can hinder the communication and validation of simulations. The Model-Driven Engineering (MDE) of simulations has been proposed in other fields to address these issues. Such approaches develop graphical Modelling Languages (MLs) that researchers use to model their problems, and then semi-automatically generate simulations from those models. Working in this way promotes communication, platform independence, incremental development, and reutilisation. This paper presents the first steps for a MDE framework for traffic simulations. It introduces a tailored extensible ML for domain experts. The ML is focused on human actions, so it adopts an Agent-Based Modelling perspective. Regarding traffic aspects, it includes concepts commonly found in related literature following the Driver-Vehicle-Environment model. The language is also suitable to accommodate additional theories using its extension mechanisms. The approach is supported by an infrastructure developed using Eclipse MDE projects: the ML is specified with Ecore, and a model editor and a code generator tools are provided. A case study illustrates how to develop a simulation based on a driver's behaviour theory for a specific target platform using these elements.
Multi-agent Behavior in a GIS Traffic Simulation
bianca.cs.trinity.edu
This paper examines the problem of traffic congestion at a microscopic level. We propose a way to understand traffic by building a simulation model that imitates human driver behavior. Our implementation has a heavy focus on unique driver characteristics and non-normative ...
A Versatile Multi-Agent Traffic Simulator Framework Based on Real Data
International Journal on Artificial Intelligence Tools, 2016
Among real-system applications of AI, the field of traffic simulation makes use of a wide range of techniques and algorithms. Especially, microscopic models of road traffic have been expanding for several years. Indeed, Multi-Agent Systems provide the capability of modeling the very diversity of individual behaviors. Several professional tools provide comprehensive sets of ready-made, accurate behaviors for several kinds of vehicles. The price in such tools is the difficulty to modify the nature of programmed behaviors, and the specialization in a single purpose, e.g. either studying resulting ows, or providing an immersive virtual reality environment. Thus, we advocate for a more exible approach for the design of multi-purpose tools for decision support. Especially, the use of geographical open databases offers the opportunity to design agent-based traffic simulators which can be continuously informed of changes in traffic conditions. Our proposal also makes decision support system...
ARCHISIM: a behavioral multi-actors traffic simulation model for the study of a traffic system including ITS aspects
HAL (Le Centre pour la Communication Scientifique Directe), 2007
Traffic phenomena come on the one hand from supply / demand mechanisms and on the other hand from the interactions between the various actors involved. Simulation models have been developed for several decades by traffic engineers to reproduce the phenomena. Based on the identification of observed traffic, they are unfortunately limited when the study is related to future situations (i.e. non existing, thus non observable, ones). Driver models have also been developed for decades by psychologists, but these models are also often very limited (i.e. they deal with very few and very specific driving tasks) and not operational (i.e. they are conceptual models). The simulation of the impact of a change in the traffic system is nevertheless a key issue, both from the safety and the capacity standpoints. The behaviour of drivers facing a new situation is extremely difficult to forecast, since human beings easily adapt their behaviour in response to infrastructure and equipments. They will not always use them according to designers' expectations (a rational use for collective optimisation) but, on the contrary, they very often follow individual issues, such as minimisation of constraints or economy of manoeuvres. These different standpoints often lead to incoherences between design and uses, which have a negative impact on safety as well as on capacity. Designing tools allowing a systemic approach of changes in the traffic system is the main objective of the INRETS MSIS department. Based on the joint use of a driving simulator and behavioural traffic simulation, the proposed approach (called "integrated approach") consists of a four stage iterative process which jointly uses a driving simulator and a behavioural microscopic traffic simulation model. To carry on studies according to this approach, MSIS team has designed a behavioural traffic simulation model and a driving simulator architecture, both novel.
Evolving Individual Behavior in a Multi-agent Traffic Simulator
2010
In this paper, we illustrate the use of evolutionary agents in a multi-agent system designed to describe the behavior of car drivers. Each agent has the selfish objective to reach its destination in the shortest time possible, and a preference in terms of paths to take, based on the presence of other agents and on the width of the roads. Those parameters are changed with an evolutionary strategy, to mimic the adaptation of a human driver to different traffic conditions. The system proposed is then tested by giving the agents the ability to perceive the presence of other agents in a given radius. Experimental results show that knowing the position of all the car drivers in the map leads the agents to obtain a better performance, thanks to the evolution of their behavior. Even the system as a whole gains some benefits from the evolution of the agents’ individual choices.
Lane Changing Behavior in a Multi-agent Traffic Simulation
bianca.cs.trinity.edu
This paper examines the problem of traffic congestion at a microscopic level. We propose a way to understand traffic by building a simulation model that imitates human driver behavior. Our implementation has a heavy focus on unique driver characteristics and non-normative ...
ARCHISIM: a behavioural multi-actors traffic simulation model for the study of a traffic system including ITS aspects
2007
Traffic phenomena come on the one hand from supply / demand mechanisms and on the other hand from the interactions between the various actors involved. Simulation models have been developed for several decades by traffic engineers to reproduce the phenomena. Based on the identification of observed traffic, they are unfortunately limited when the study is related to future situations (i.e. non existing, thus non observable, ones). Driver models have also been developed for decades by psychologists, but these models are also often very limited (i.e. they deal with very few and very specific driving tasks) and not operational (i.e. they are conceptual models). The simulation of the impact of a change in the traffic system is nevertheless a key issue, both from the safety and the capacity standpoints. The behaviour of drivers facing a new situation is extremely difficult to forecast, since human beings easily adapt their behaviour in response to infrastructure and equipments. They will not always use them according to designers' expectations (a rational use for collective optimisation) but, on the contrary, they very often follow individual issues, such as minimisation of constraints or economy of manoeuvres. These different standpoints often lead to incoherences between design and uses, which have a negative impact on safety as well as on capacity. Designing tools allowing a systemic approach of changes in the traffic system is the main objective of the INRETS MSIS department. Based on the joint use of a driving simulator and behavioural traffic simulation, the proposed approach (called "integrated approach") consists of a four stage iterative process which jointly uses a driving simulator and a behavioural microscopic traffic simulation model. To carry on studies according to this approach, MSIS team has designed a behavioural traffic simulation model and a driving simulator architecture, both novel.
A Model-Driven Engineering Process for Agent-based Traffic Simulations
2015
Traffic has an important impact in many aspects of our everyday life, from healthcare to transport regulation or urban planning. Given its complexity, the study in real settings is frequently limited, so researchers resort to simulations. However, realistic simulations are still complex systems. Its development frequently requires multidisciplinary groups, where misunderstandings are frequent, and there is a great variety of potential theories and platforms to consider. In order to reduce the impact of these issues, the Model-Driven Engineering (MDE) of simulations has been proposed. It is focused on developing mainly through models and their semi-automated transformation. Nevertheless, an effective approach of this kind requires the availability of infrastructures that include modelling languages, transformations, tools, and processes to use them. This work presents a MDE process for traffic simulations. It introduces a modelling language and makes uses of available infrastructures in its tasks. The process guides users in creating tailored models for their simulations, and transforming these to code. A case study that uses an existing model for drivers' behaviour and an already available platform to develop a simulation illustrates the approach.
Multilayered Multiagent System for Traffic Simulation: (Demonstration)
2016
We propose a multilayered multiagent simulator that can simulate traffic in any urban environment on earth, subject to specific weather conditions. We adopt an agent-based approach for the behaviors of the vehicles and the drivers. We additionally propose a behavioral model to realistically emulate the driving behaviors of humans.
Multi-Agent Based Simulation of Individual Traffic in Berlin
2005
Multi-agent simulations of traffic are widely expected to become an important tool for transportation planning in the mid-term future. This paper reports on the first steps of a project which aims to apply such a tool to a large real world scenario based on datasets create in the normal transportation planning process for use with established transportation planning tools. As the first steps of the implementation show, many problems related to different data semantics and the different modelling concepts can occur. In most cases, theses problems can be resolved by minor adoptions of the software or the data. The evaluation of a large scenario of several hundred thousand agents shows that performance issues do no longer hinder applications of this kind. Thereby, the implementation of this scenario helps to push multi-agent traffic simulation tools forward to real world applications.
TrafficGen: A Flexible Tool For Informing Agent-Based Traffic Simulations With Open Data
Lecture Notes in Computer Science, 2015
Multi-Agent Systems have been widely used for traffic simulation. The modeling of individuals allows indeed to introduce a behavioral diversity which is crucial to obtain realistic simulation outcomes. The recent growth of open geographical databases and related flow information provides an opportunity for enhancing traffic simulators with data automatically retrieved from the real world and updated regularly. We present here TrafficGen, a highly modular platform based on the integration of such open data within a library of rule-based behaviors, in order to provide a versatile decision support tool in traffic.
Reconciling Strategic and Tactical Decision Making in Agent-Oriented Simulation of Vehicles in Urban Traffic
2011
We consider an integrated decision making process of autonomous vehicles in agent-oriented simulation of urban traffic systems. In our approach, the planning process for a vehicle agent is separated into two stages: strategic planning and tactical planning. During the strategic planning stage the vehicle agents constructs the optimal route from source to destination; during the tactical planning stage the operative decisions such as speed regulation and lane change are considered. For strategic planning we modify the stochastic shortest path algorithm with imperfect knowledge about network conditions. For tactical planning we apply distributed multiagent reinforcement learning with other vehicles at the same edge. We present planning algorithms for both stages and demonstrate interconnections between them; an example illustrates how the proposed approach may reduce travel time of vehicle agents in urban traffic.
Traffic generation with the SCANeR II simulator : towards a multi-agent architecture
Renault has been developing driving simulators for over 10 years. They are used for ergonomics and advanced engineering studies at Renault as well as for road traffic research, human factor studies and driver training in different labs and companies in Europe (especially in France, the UK, Sweden and Norway). These simulators use SGI and/or PC image generation technology with up to 6 graphics channels and motion seat or platforms for kinaesthetic rendering. The real time simulation software, SCANeR II version 1.3, currently includes -among otherssimulation session initialisation and monitoring, vehicle dynamics, traffic generation, visual and kinaesthetic modules. The traffic generation software allows the user (i.e. the experimenter) to describe and initiate real time traffic with autonomous vehicles. Up to several hundred vehicles can be rendered using a traffic management system and several state machines standing for the different vehicles. The SCANeR II software has already bee...
An Agent-Based Simulation Model for Urban Traffic System
Computer and Information Science, 2011
In this paper, we propose a model for vehicle traffic based on multi-agent systems and account suppositions and its issues. Traffic is an ever-growing problem as population and the number of drivers around the world increase exponentially. Previously, fluid flow models have been used in an attempt to model traffic. Based on recent studies, only agent based models can accurately model a traffic scenario. This is because small perturbations could have a butterfly-like effect, which causes a rapid change in the entire system.
Multi-agent Architecture for Simulation of Traffic with Communications
Inter-vehicle communications, in the context of Intelligent Transportation Systems, will probably bring a significant improvement in both traffic safety and efficiency. In order to evaluate in what measure this is true, traffic simulations that take into account the communications between vehicles are needed. In this paper, we propose an agent-based architecture, in which the simulation and management of the intervehicle communications are integrated in the simulation of vehicles, in a hierarchical multi-agent environment. An overview of multi-agent methodologies, platforms, among other, is also presented.
Microscopic traffic simulation with reactive driving agents
ITSC 2001. 2001 IEEE Intelligent Transportation Systems. Proceedings (Cat. No.01TH8585)
Computer traffic simulation is important for making new traffic-control strategies. Microscopic traffic simulators can model traffic flow in a realistic manner and are ideal for agent-based vehicle control. In this paper we describe a model of a reactive agent that is used to control a simulated vehicle. The agent is capable of tactical-level driving and has different driving styles. To ensure fast reaction times, the agent's driving task is divided in several competing and reactive behavior rules. The agent is implemented in and tested with a prototype traffic simulator program. The simulator consists of an urban environment with multi-lane roads, intersections, traffic lights, and vehicles. Every vehicle is controlled by a driving agent and all agents have individual behavior settings. Preliminary experiments have shown that the agents exhibit human-like behavior ranging from slow and careful to fast and aggressive driving behavior.
Related topics
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.