AMIRA - Advanced Multimodal Intelligence for Remote Assistance

The business goal of AMIRA (Advanced Multimodal Intelligence for Remote Assistance) was to significantly improve the accessibility and resources available to support urgent and critical diagnostics and decisions that must be taken by mobile workers, operating individually or in multi-discipline collaborations, at their point of intervention in an event.

The technical goal of AMIRA was to develop a set of reusable components using search, reasoning, speech dialogue technology and collaborative working techniques that could be used to create a variety of applications for use by mobile workers operating in safety or business critical situations in the field. The prototype application developed in AMIRA is a wireless, easy to use, intelligent, real-time diagnostic and decision support system for mobile workers using multi-modal devices such as wireless earphone/microphone, PDA or UMTS phone. Although the envisaged application is capable of being used across a diverse range of disciplines and sectors, the focus of the work, and the basis for this project, was on proving the application in providing Multi-modal assistance at the point of intervention in time-critical, safety/business critical incidents. In this way, AMIRA was demonstrated as providing a tool for the mobile worker in both fairly simple and highly complex scenarios.

Building on the work and results of current and previous RTD projects, AMIRA widened the applicability of advanced search and reasoning technologies to make it possible to access these through a speech dialogue interface. By coupling speech dialogue technologies with search and reasoning technologies, structural CaseBased Reasoning (CBR) drove the intelligence behind the dialogue, generating the questions to be automatically answered (in the same way as it does today for Web self service but targeting speech self service). AMIRA provided semantically based and context aware systems that can acquire, organise, share and reuse knowledge in structured and unstructured format. It extended CBR with the ability to process trends in data, in order to be able to identify and recall cases with similar parameter shifts.

Proof-of-concept trials were conducted in safety-critical situations (through the Fire Service College and operational fire services) and business-critical situations (through operations of West Midlands Fire Services - Transport Engineering Workshops). The proof-of-concept application used a spoken dialogue front-end to capture the operatives initial description of the symptoms. The interactive system extracted semantic descriptions from the operatives spoken dialogue and interactively constructs a problem description. The problem description was forwarded to an integrated reasoning and search system. This backend system scanned its databases for cases matching the problem description. If necessary it returned a request for additional information to the interactive system in order to narrow down a too large set of possibly matching cases. The interactive system elicited this information from the operative.

Once the set of matching cases was sufficiently defined, the back-end system retrieved, extracted or inferred a sequence of action descriptions that helped the operative to fully identify the source of the problem and provided best practice support to support the operative resolve the problem.

Funding

The AMIRA project was funded by the European Commission under grant number FP6, project IST-2003-511740. The project was carried out between July 2004 and June 2006.

Consortium

The Consortium consists of seven partners.

Research Team

Six members of the Wi2-group are working or have been working within AMIRA.

Publications

2007

  • Andrea Freßmann, Ralph Bergmann, Brian Taylor, Ben Diamond, and Gary Carr-Smith. Mobile Knowledge Management Support in Fire Service Organisations. In Tagungsband der 8. Internationalen Tagung der Wirtschaftsinformatik, 2007, Karlsruhe University Press (accepted to be published)

2006

  • Andrea Freßmann. Adaptive Workflow Support for Search Processes within Fire Service Organisations. In Sumitra Mitra Reddy, editor, Proceedings of the fifteenth IEEE International Workshop on Enabling Technologies: Infrastructure for Collaborative Enterprises, pages 291-296, June 2006, IEEE Computer Society.
  • Ralph Bergmann, Andrea Freßmann, Kerstin Maximini, Rainer Maximini, and Thomas Sauer. Case-Based Support for Collaborative Business. In Advances in Case-Based Reasoning - Proceedings of the 8th European Conference, ECCBR 2006, Fethiye, Turkey, volume 4106 of LNCS, pages 519-533, September 2006, Springer Verlag, Berlin-Heidelberg.

2005

  •   Eric Auriol, Paul Heisterkamp, Ingvar Aarberg, Ben Diamond, Brian Taylor, and Andrea Freßmann. AMIRA: Supporting Diagnostics And Decisions In Safety-Critical Incidents. In Paul Cunningham, and Miriam Cunningham, editors, Innovation and the Knowledge Economy: Issues, Applications, Case Studies, 2005, 2005 IOS Press Amsterdam.
  • Andrea Freßmann, Rainer Maximini, and Thomas Sauer. Towards Collaborative Agent-Based Knowledge Support for Time-Critical and Business-Critical Processes. In Klaus-Dieter Althoff, Andreas Dengel, Ralph Bergmann, Markus Nick, and Thomas Roth-Berghofer, editors, Professional Knowledge Management: Third Biennial Conference, WM 2005, Kaiserslautern, Germany, April 10-13, 2005, Revised Selected Papers, volume 3782 of LNAI, pages 420-430, December 2005, Springer-Verlag GmbH.
  • Andrea Freßmann, Thomas Sauer, and Ralph Bergmann. Collaboration Patterns for Adaptive Software Engineering Processes. In Hans Czap, R. Unland, C. Branki, and H. Tianfield, editors, Self-Organization and Autonomic Informatics (I), volume 135, pages 304-312, Amsterdam, Netherlands, November 2005, Frontiers in Artificial Intelligence and Applications (FAIA), IOS Press.
  • Andrea Freßmann, Kerstin Maximini, Rainer Maximini, and Thomas Sauer. Collaborative Agent-based Knowledge Support for Empirical and Knowledge-intense Processes. In MATES 2005 / CIA 2005, volume 3550 of LNAI, pages 235-236, Koblenz, Germany, September 2005, Springer-Verlag.
  • Andrea Freßmann, Kerstin Maximini, Rainer Maximini, and Thomas Sauer. CBR-based Execution and Planning Support for Collaborative Workflows. In Workshop "Similarities - Processes - Workflows" on the Sixth International Conference on Case-Based Reasoning (ICCBR 2005), pages 271-280, Chicago, Illinois (USA), August 2005.
  • Rainer Maximini, Andrea Freßmann, Thomas Sauer, Kerstin Maximini, and Ralph Bergmann. CAKE - Collaborative Agent-based Knowledge Engine. Technical Report, University of Trier, Department of Business Information Systems II, July 2005.
  • Andrea Freßmann, Rainer Maximini, and Thomas Sauer. Towards Collaborative Agent-based Knowledge Support for Agile Projects. In Klaus-Dieter Althoff, Andreas Dengel, Ralph Bergmann, Markus Nick, and Thomas Roth-Berghofer, editors, WM2005: Professional Knowledge Management Experiences and Visions, pages 383-388, Kaiserslautern, April 2005, German Research Center for Artificial Intelligence DFKI GmbH.
  • Andrea Freßmann and Fawsy Bendeck. D2.1.0 - Socio-economic Study of User Needs. Technical Report, University of Trier, Department of Business Information Systems II, March 2005.

 2004

  •  Rouven Thimm and Thomas Will. AMIRA - A State of the Art Analysis of Workflow Modelling and Agent-based Information-Systems. Technical Report, University of Trier, Department of Business Information Systems II, October 2004.