Design for Failure: Intelligent Systems Learning from Their Mistakes

Smart environments aim to make the life of their inhabitants more comfortable by having context-aware systems continuously work together to assist people with their daily tasks. However, all too often these assistive technologies are naively or optimistically developed assuming that systems can always anticipate what users want. Furthermore, the more these smart systems grow in complexity, the more prone to failure they become. The overall goal of this paper is to define new concepts and methodologies for the development of more reliable smart applications, and propose middleware support to analyze failures in context-aware behavior, culminating in a software-based safeguard that improves robustness against unforeseen human interventions, exceptional circumstances and unexpected events.

[1]  Juan Carlos Augusto,et al.  Software simulation and verification to increase the reliability of Intelligent Environments , 2013, Adv. Eng. Softw..

[2]  Geoff Hulten,et al.  Mining high-speed data streams , 2000, KDD '00.

[3]  Diane J. Cook,et al.  Human Activity Recognition and Pattern Discovery , 2010, IEEE Pervasive Computing.

[4]  Cecilia Mascolo,et al.  Mobile Computing Middleware , 2002, NETWORKING Tutorials.

[5]  Yolande Berbers,et al.  SAMURAI: A Streaming Multi-tenant Context-Management Architecture for Intelligent and Scalable Internet of Things Applications , 2014, 2014 International Conference on Intelligent Environments.

[6]  Yolande Berbers,et al.  Learning Deployment Trade-offs for Self-Optimization of Internet of Things Applications , 2013, ICAC.

[7]  David Luckham,et al.  The power of events - an introduction to complex event processing in distributed enterprise systems , 2002, RuleML.

[8]  Euiho Suh,et al.  Context-aware systems: A literature review and classification , 2009, Expert Syst. Appl..

[9]  Wil M. P. van der Aalst,et al.  Workflow mining: discovering process models from event logs , 2004, IEEE Transactions on Knowledge and Data Engineering.

[10]  Juan Carlos Augusto,et al.  Living without a safety net in an Intelligent Environment , 2011, EAI Endorsed Trans. Ambient Syst..

[11]  Diane J. Cook,et al.  Discovering frequent user--environment interactions in intelligent environments , 2011, Personal and Ubiquitous Computing.

[12]  David Wetherall,et al.  Recognizing daily activities with RFID-based sensors , 2009, UbiComp.

[13]  John Allspaw,et al.  How Complex Systems Fail , 2010, Web Operations.

[14]  Geoff Holmes,et al.  CD-MOA: Change Detection Framework for Massive Online Analysis , 2013, IDA.

[15]  Yolande Berbers,et al.  Consistency in Context-Aware Behavior: a Model Checking Approach , 2012, Intelligent Environments.

[16]  Jadwiga Indulska,et al.  Middleware for Distributed Context-Aware Systems , 2005, OTM Conferences.

[17]  Wen Yao,et al.  Leveraging complex event processing for smart hospitals using RFID , 2011, J. Netw. Comput. Appl..

[18]  Alfons Schuster.  Robust Intelligent Systems , 2008 .