A Tool for the Creation of HLA Compliant Training Simulations

A Tool for the Creation of HLA Compliant Training Simulations Rafaela Vilela da Rocha1, Regina Borges de Araújo1, Azzedine Boukerche2 1 Departamento de Computação Universidade Federal de São Carlos - UFSCar São Carlos-SP, Brasil 2 SITE University of Ottawa Ottawa, Canada Some of the applications being built as part of the INCT-SEC Project involve security and safety of critical infrastructures and national borders. Monitoring and surveillance will be done with the integration of aerial and terrestrial non-tripulated vehicles and wireless sensor networks. When emergency situations are spotted, emergency management teams might be sent to deal with risky situations. Virtual environments for training simulations introduce a safe environment to investigate human behavior and response in dangerous situations, reducing real risks to life and assets. These complex environments are difficult to build, control and manage. This paper presents an architecture to support the modeling of different training simulations. This architecture is based on the IEEE standard High Level Architecture for interoperability and reusability of distributed simulation. The architecture integrates multiple ontologies through the concept of nonlinear interactive storytelling to support easier simulation instantiation Simulations can be accessed through any device that can run an X3D browser. A case study on the training of emergency management teams is described. Laboratories: WINDIS, Paradise Publication: DS-RT 2009, CLEI 2010 Funding: FAPESP, CNPq, CAPES, Ministério da Ciência e Tencologia Science Highlights Training simulations are models of existing or proposed systems such as typical simulation models. However, the decision-making on resources and operational policies is made during the simulation run and the outputs are also observed during it. This allows the users to see the impact of their decisions on the simulation environment in real-time [1]. Virtual environments for training simulations introduce a safe environment to investigate human behavior and response in dangerous situations, reducing real risks to lives and assets. However, most of the existing Collaborative Virtual Environments (CVEs) supporting systems are tuned to specific tasks and their architectures are, typically, tightly coupled to the applications [2]. This makes any modification to the application dependent on programming, turning large scale CVEs construction and extension a challenging task. Desirable and challenging characteristics for simulation environments include: modularity, reuse, hierarchical structure of the model, scalability, portability, interoperability, distribution execution, execution over the internet, and ease of use [3]. This work presents a modular architecture to support the development, implementation, management, control and analysis of collaborative virtual environments for training simulations. A tool is also presented, which uses the concept of Nonlinear Interactive Storytelling to integrate a set of ontologies to create different simulation scenarios. The training simulations architecture described in this paper is based on the IEEE standard High Level Architecture (HLA) and Federation Development and Execution Process (FEDEP) [4][5] for interoperability and reusability of distributed simulation models. HLA B ASED S IMULATIONS The HLA has three main components: Framework and Rules, Object Model Template Specification and Federate Interface Specification. The Framework and Rules describe the responsibilities federates (simulations, supporting utilities or interfaces to live systems) and federations (set of federates working together) must hold for implementation [4]. The Object Model Template (OMT) Specification defines the Federation Object Model (FOM) and the Management Object Model (MOM) [4]. The FOM describes all the data exchanged among the federates in a common standardized format. The MOM is a set of management objects and interactions classes that can be included in a FOM and is used to extract information on the executing federation and its participating federates. The Federate Interface Specification defines the standard services and interfaces for the software architecture called Runtime Infrastructure (RTI) to support the data exchange among the federates [4]. These interfaces are arranged in seven service groups: federation management, declaration management, object management, ownership management, time management, data distribution management and support services. RTI is a distributed operational system that offer services to interaction among federates and to federation management. The Federation Development and Execution Process (FEDEP) are recommended practices to develop and execute federations [5]. These processes and procedures were used to define the steps to develop the simulation model in our architecture. In this work, simulation construction is driven by nonlinear, interactive storytelling and instantiated ontologies. The story is nonlinear because it allows the trainer/trainees to influence future events and change the direction of the story [6]. Ontology provides a common domain vocabulary, usually formal enough to support automatic inference and ideally requires a community consensus [7][8]. It serves as a basis for communication, interoperability, reusability, reliability and system specification [9]. Ontologies have been acknowledged as a potential concept for simulation modeling and analysis [10][11][12]. AN ARCHITECTURE TO SUPPORT THE CREATION OF EMERGENCY MANAGEMENT TRAINING SIMULATIONS In order to help non-computer specialists to create HLA compliant training simulation, the architecture presented in Figure 1 was devised and its main components are described as follows. 1) Simulation model development In the stage of simulation model development, the trainer can create, instantiate and integrate the ontologies, as well as automatically generate the FDD file and test the simulation. He/She can use the Simulations Creation Tool and access the databases (ontologies, FDD DB, X3D DB and Context DB). Simulation Creation Tool provides for trainers (training commander) a graphical interface to create different simulation scenarios. These scenarios can be instantiated through the use of a set of ontologies and a context database with information collected and interpreted from physical environments, through the concept of nonlinear interactive storytelling. The trainer can use existing 3D geometric models (building, persons and objects), which are in an X3D BD, to create a particular virtual worlds for his/her trainees. Seven ontologies were identified as necessary for emergency training simulation modeling: simulation, infrastructure, person, object, vehicle, emergency and tactic described in [13]. As shown in Figure 2, the ontologies are integrated to define the scenery, characters and objects of virtual environments, as well as the events that can occur during the simulation. Science Highlights Figure 1. Overview of the architecture to support the creation of training simulations. The ontologies were described in Web Ontology Language (OWL) - a language recommended by the World Wide Web Consortium (W3C) to build its [14][15]. The ontologies creation and integration were made with the Protégé-OWL Editor - tool to create and edit an ontology using a graphical interface [16] A Federation Document Data database (FDD DB) contains FDD files generated by the Simulation Creation Tool for an instantiated simulation model (i.e., an HLA federation). Figure 2. Integrated ontologies for training simulation creation. A 3D geometric model database (X3D DB) contains X3D 1 model used to build the geometric worlds used as scenarios for the simulation applications (emergency management for critical cyberinfrastructures). A context database (Context DB) contains data collected from physical environments by means of Wireless Sensor Network integrated to Radio Frequency IDs. This data is collected and interpreted by services in a Middleware (MidSensorNet- service middleware for the emergency management domain), described in [17]. This data is stored in different interpretation levels: from aggregated data captured by sensors from the physical environment to incident information interpreted by a context interpretation service. 2) Simulation model execution In the stage of executing simulation model, the trainer can create a federation from the generated simulation model in the previous stage and HLA Compliant Simulation Repository, using the Simulation Controller / Monitor module. Trainees can test their skills using the X3D Visualizer module. HLA Complaint Simulations Repository contains HLA complaint simulations, which control the objects added to the virtual environment. These simulations are integrated with open source engines, such as physical and human body simulation engines to provide more realistic virtual environments. 1 X3D (Extensible 3D): is an ISO standard for real-time 3D graphics that defines a file format using XML and a run-time environment to be embedded in applications. Science Highlights Simulation Controller/ Monitor is responsible for creating the federation from ontologies and FDD file. It is an X3D Visualizer with one additional functionality to create/ destroy federations. X3D Visualizer is responsible for the visualization of the simulation. The X3D Visualizer, built with open standard and web-based framework, generates HLA compliant simulation visualization interfaces with support to control and management functionalities. This framework, described in [18], uses open-source software and is designed to make it easier to implement and deploy visualization applications on different platforms with customizable interface for simulations runtime control and management. The special characteristic of our X3D Visualizer is its support to simulation control and management. In each X3D Visualizer, a trainee can control his/her avatar, and make his/her own decisions. Every action on the visualizer is logged for later analysis. Federates of the same federation communicate with each other through the RTI infrastructure. 3) Output Analyzis In the stage of analyzing simulation model, the trainer can view and generate performance reports of all trainees using the Simulation Output Analyzer Tool. This tool access Output Repository and generate these reports. Output Repository contains data resulting from simulation execution. It contains all the information about trainees’ performance. Generated outputs can be used for future evaluation. Simulation Output Analyzer Tool is a software tool to read the simulation outputs and to generate reports for the training commanders. It supports the strategies creation to combat the emergency, which enables verification and validation of the activities carried out by trainees. Simulation Output Analyzer Tool is a software tool to read the simulation outputs and to generate reports for the training commanders. CASE STUDY For our proof of concept an industrial plant was used, which presents potential for fire accidents: a fry chip factory (any other emergency environment could be used – with corresponding 3D model). The infrastructure X3D model was created for the simulation. 3D fire fighters and fire trucks were downloaded from the ©Google SketchUp 3D Warehouse and converted to X3D files (all geometric models are kept in X3D BD). Figure 3 illustrates the case study. All information for the training simulation is added (instantiated) and / or selected with the Simulation Creation Tool. Initially, the infrastructure ontology was instantiated with the plan details. Soon after, it was instantiated the person ontology with all Fire Department staff that will participate in the training, together with the instantiation of their equipment and fire fighting vehicles that can be used. Figure 3. X3D building plan of the case study. The emergency ontology was instantiated with initial conditions required for a fire (specific fuel, ignition source and oxidizing agent). In our proof of concept, the emergency is a fire, having as its source the production place. The fire is generated by the combination of fuel (inflammable solid – the packing-cases in the production), oxidizing agent (oxygen) and ignition source (short-circuit). Finally, the simulation ontology was instantiated. It incorporates all six other ontologies for a specific simulation model. The training goal, the scenario (infrastructure that is used and emergencies that occur), the characters (rescue team and victims), the fighting strategies (which are the tactics to be employed in emergency events that may occur and also the mapping of attack lines and vehicles position on the plan) and other events that can be scheduled to start during the simulation, with prerequisites, are instantiated in the simulation ontology. All instantiations mentioned were successfully realized through our set of ontologies. The Simulation Creation Tool generates FDD file for the HLA federation and this simulation can be created by the Simulation Controller / Monitor. The federates are added in the federation by the X3D Visualizer. CONCLUSION Systems that maximize emergency management team’s training to prepare and to respond emergencies cannot cease accidents to occur. However, they can potentially improve the teams’ work quality and reduce the accidents during a real occurrence. This paper described both an architecture and a tool to create HLA compliant training simulations. The integration of multiple ontologies is used to generate different simulations scenarios. Ontologies were developed that follow standards used in São Paulo State (Brazil) for protection against fire. Simulation control is being implemented as part of our tool. A simulation output analyzer is being developed and will be integrated to the simulation creation tool. Accurate monitoring of persons will be implemented as future work. Science Highlights ACKNOWLEDGEMENTS The authors thank CNPq and FAPESP for the support to the INCT-SEC Project, processes 573963/2008-8 e 08/57870-9. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] C.A. Chung. “Simulation Modeling Handbook: A Practical Approach”. CRC Press, 2004. A. Boukerche, R. B. Araújo, D. D. Duarte, and L. Andrade. “A Novel Solution for the Development of Collaborative Virtual Environment Simulations in Large Scale”. In: IEEE Proceedings of the The 9-th International Symposium on Distributed Simulation and Real Time Applications, 2005, vol. 9, pp. 1-8. E.M. Abu-taieh and A.A. Sheikh. Handbook of Research on Discrete Event Simulation Environments: technologies and applications, Hershey, New York: IGI Global, 2010. IEEE Std. 1516-2000. “IEEE Standard for Modeling and Simulation (M&S) High Level Architecture (HLA) – Framework and Rules”, 2001. Available em: http://www.ieee.org/, 2009. IEEE Std 1516.3-2000. “IEEE Recommended Practice for High Level Architecture (HLA) Federation Development and Execution Process (FEDEP)”. New York: Institute of Electrical and Electronics Engineers, 2003. E. Adams and A. Rollings. “Game Design and Development”. Person Education, Inc., Upper Saddle River, New Jersey, 2007. G. Pérez. “A. Ontological Engineering: A state of the art”. Expert Update. British Comp.Society, 1999, vol. 2, n. 3, pp. 33-43. R. Vieira, D.S. Abdalla, D. M. Silva, and M. R. Santana. “Web Semântica: Ontologias, Lógica de Descrição e Inferência”. In: Web e Multimidia: Desafios e Soluções (WebMedia 2005 - Minicursos). 1 ed. Porto Alegre: SBC, 2005, vol. 1, pp. 127-167. M. Uschold and M. Gruninger. “Ontologies: Principles, methods and applications”. Knowledge Engineering Review, vol. 11, n. 2, pp. 93-155, February 1996. P. Benjamin, M. Patki and R. Mayer. Using Ontologies for Simulation Modeling, Proc. of the 2006 Winter Simulation Conference, Monterey, California, USA, December 3-6, 2006. J.A. Miller, G.T. Baramidze, A.P. Sheth, and P.A. Fishwick. Investigating Ontologies for Simulation Modeling, Proceedings of the 37th annual symposium on Simulation, 2004. G. Silver, L.Lacy and J. Miller. Ontology Based Representations of Simulation Models Following the Process Interaction World View. Proc. of the 2006 Winter Simulation Conf. Monterey, California, USA, December 3-6, 2006. R.V. Rocha, R.B. Araújo, M.R. Campos, and A. Boukerche. "HLA Compliant Training Simulations Creation Tool," Proceedings of the 2009 13th IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications, Singapore: IEEE Computer Society, 2009, p. 192–198. M. K. Smith, C. Welty, and D.L. McGuinness. “OWL Web Ontology Language Guide”. W3C Recommendation 10 February 2004. Available em: http://www.w3.org/TR/2004/REC-owl-guide-20040210/, 2008. S. Bechhofer, F. Harmelen, J. Hendler, I. Horrocks, D.L. McGuinness, P.F. Patel-Schneider, and L.A. Stein. “OWL Web Ontology Language Reference”. W3C Recommendation 10 February 2004. Available em: http://www.w3.org/TR/2004/REC-owl-ref-20040210/, 2008. [16] Protégé. Stanford Center for Biomedical Informatics Research, 2009. Available em: http://protege.stanford.edu/, 2009. [17] R.B. Araújo, L.A. Villas, and J.E. Ribeiro. “MidSensorNet: A Service Middleware to Emergency Management”. Internal Technical Report, February, 2008 (in Portuguese). [18] R.B. Araújo, F.M. Iwasaki, E. Pizzolato, and A. Boukerche. “A Framework for 3D Web-Based Visualization of HLACompliant Simulations”. Web3D 2008: Infrastructure Evolution. 13th International Symposium on 3D Web Technology - Co-located with SIGGRAPH 2008. Los Angeles, August 2008.