Jacques Tisseau

GVT: a platform to create virtual environments for procedural training

The use of virtual environments for training is strongly stimulated by important needs for training on sensitive equipments. Yet, developing such an application is often done without reusing existing components, which requires a huge amount of time. We present in this paper a full authoring platform to facilitate the development of both new virtual environments and pedagogical information for procedural training. This platform, named GVT (generic virtual training) relies on innovative models and provides authoring tools which allow capitalizing on the developments realized. We present a generic model named STORM, used to describe reusable behaviors for 3D objects and reusable interactions between those objects. We also present a scenario language named LORA which allows non computer scientists to author various and complex sequences of tasks in a virtual scene. Based on those models, as an industrial validation with Nexter-Group, more than fifty operational scenarios of maintenance training on military equipments have been realized so far. We have also set up an assessment campaign, and we expose in this paper the first results which show that GVT enables trainees to learn procedures efficiently. The platform keeps on evolving and training on collaborative procedures will soon be available.

An immune oriented multi-agent system for biological image processing

Abstract In this article, we present a parallel image processing system based on the concept of reactive agents. Our system lies in the oRis language, which allows to describe finely and simply the agents’ behaviors to detect image features. We also present a method of segmentation using a multi-agent system, and two biological applications made with oRis. The stopping of this multi-agent system is implemented through a technique issued from immunology: the apoptosis.

Transfer of learning in virtual environments: a new challenge?

The aim of all education is to apply what we learn in different contexts and to recognise and extend this learning to new situations. Virtual learning environments can be used to build skills. Recent research in cognitive psychology and education has shown that acquisitions are linked to the initial context. This provides a challenge for virtual reality in education or training. A brief overview of transfer issues highlights five main ideas: (1) the type of transfer enables the virtual environment (VE) to be classified according to what is learned; (2) the transfer process can create conditions within the VE to facilitate transfer of learning; (3) specific features of VR must match and comply with transfer of learning; (4) transfer can be used to assess a VE’s effectiveness; and (5) future research on transfer of learning must examine the singular context of learning. This paper discusses how new perspectives in cognitive psychology influence and promote transfer of learning through the use of VEs.

ARéVi: A Virtual Reality Multiagent Platform

AReVi (in French, Atelier de Realite Virtuelle) is a distributed virtual reality toolkit. Its kernel (a group of C++ classes) makes it possible to create cooperative and distributed virtual reality applications by minimizing the programming effort. AReVi is built around a dynamic multiagent language: oRis. At any time, this language allows to stop the ongoing session, to add new entities (known or not when starting the session), to modify an entity or an entire entity family behavior. More generally, oRis gives the user the ability to interact with the agents by directly using their language, thus offering a way of immersion through the language.

Educative distributed virtual environments for children

This paper presents a distributed virtual reality environment for children. This virtual environment supports cooperation among members of a dispersed team engaged in a concurrent context. The virtual environment maintains a shared information space described in a standard Virtual Reality Modeling Language (VRML) format. Users are allowed to interact and to give decisions using cooperative mechanisms. A user-friendly interface enables teachers to create their own stories that fit with children pedagogical requirements and generates new virtual environments according to teachers specifications. The implementation is based on DeepMatrix as environment server, VRML and Java as languages and Cortona VRML plug-in from ParallelGraphics. It is actually running on the Internet: http://www.enib.fr/eve.

Gesture recognition in flow based on PCA and using multiagent system

In our context of Virtual Theater, a virtual actor performs with a real actor. They communicate through movements and choreography. The system has to interpret the real actors gesture into a symbolic representation. Therefore, we present a method for real-time recognition. We use properties from Principal Component Analysis (PCA) to create signature for each gesture and a multiagent system to perform the recognition.

Evaluation of procedural learning transfer from a virtual environment to a real situation: a case study on tank maintenance training

Virtual reality opens new opportunities for operator training in complex tasks. It lowers costs and has fewer constraints than traditional training. The ultimate goal of virtual training is to transfer knowledge gained in a virtual environment to an actual real-world setting. This study tested whether a maintenance procedure could be learnt equally well by virtual-environment and conventional training. Forty-two adults were divided into three equally sized groups: virtual training (GVT® [generic virtual training]), conventional training (using a real tank suspension and preparation station) and control (no training). Participants then performed the procedure individually in the real environment. Both training types (conventional and virtual) produced similar levels of performance when the procedure was carried out in real conditions. Performance level for the two trained groups was better in terms of success and time taken to complete the task, time spent consulting job instructions and number of times the instructor provided guidance. Practitioner Summary: A key issue for virtual environments for training (VETs) is the transfer of skills to real situations. An experiment investigated whether skills acquired in a VET could be applied in a real situation. Results suggest that a procedure can be successfully transferred from the virtual to the real.

A multiagent system to model an human humoral response

The immune system mechanisms are very complex and the number of parameters is extremely important. Moreover, the interactions between the different cells during an immune response induce chaos and nonlinear phenomena. Our approach consists in using the cooperative models, established by the immunologists, to build a multi-agent model. We avoid the problem of nondeterminism by only encoding the basic behaviors of the agents, and the global chaotic phenomena are induced by the interactions between the agents. The advantages are that agents can be viewed, modified, removed from the model or added to the model very easily. That is not the case in mathematics, where a modification of an assumption generally involves the rewriting of the model. This approach is possible because the behaviors of several immune system cells are known in their principal lines and some qualitative models of immune cell cooperations have been developed by immunologists. These models have already demonstrated that they are valid locally in time and in space, i.e. they do not have a global approach. With the multi-agent system, we can, thanks to the simulation, analyse the global consequences from the local behaviors and observe a qualitative striking resemblance to statistical results coming from a real experimentation. We have chosen to simulate a human secondary humoral response with a multi-agent system to study the kinetic of the antibody proliferation with several type of antigenic substances.

Multi-Agents System to Model Cell Signalling by Using Fuzzy Cognitive Maps. Application to Computer Simulation of Multiple Myeloma

In order to simulate biological processes, we use multi-agents system. However, modelling cell behavior in systems biology is complex and may be based on intracellular biochemical pathway. So, we have developed in this study a Fuzzy Influence Graph to model MAPK pathway. A Fuzzy Influence Graph is also called Fuzzy Cognitive Map.This model can be integrated in agents representing cells. Results indicate that despite individual variations, the average behavior of MAPK pathway in a cells group is close to results obtained by ordinary differential equations. So, we have also modelled multiple myeloma cells signalling by using this approach.

Fuzzy cognitive maps for the simulation of individual adaptive behaviors

This paper focuses on the simulation of behavior for autonomous entities in virtual environments. The behavior of these entities must determine their responses not only to external stimuli, but also with regard to internal states. We propose to describe such behavior using fuzzy cognitive maps (FCMs), whereby these internal states might be explicitly represented. This paper presents the use of FCMs as a tool to specify and control the behavior of individual agents. First, we describe how FCMs can be used to model behavior. We then present a learning algorithm allowing the adaptation of FCMs through observation. Copyright