Thursday January 14, 2010 3:00 – 4:00 p.m.
Mechanical Engineering – Engineering Mechanics at Michigan Technological University
This presentation introduces research activities in the Laboratory of Intelligent Mechatronicsand Embedded Systems in the Department of Mechanical Engineering Engineering Mechanicsat Michigan Tech, with a focus on the development of adaptive monitoring networks usingbiological system concepts and mechanisms. Technology is taking us to a world where numerousnetworked devices interact with the physical world in multiple ways and at multiple scales. Thefuture distributed systems will need to possess much higher quality comparing to those of todayin terms of adaptability, autonomy, and reliability due to the increased complexity of systemsand unpredictable working conditions. The fundamental research challenge is to establish robustdecentralized computing systems that interact with physical world, be capable of operating underchanging environments, and exhibit the desired response behavior under physical constraints.The biological systems are able to handle many of these challenges much more efficient thanengineered systems. Motivated by the adaptive sensing and emergent pattern recognitioncapabilities of the natural immune system, we employ an immune-inspired approach to achieveadaptive monitoring and anomaly detection. The presented approach establishes a newmonitoring paradigm by embodying desirable immune attributes, such as adaptation, immunepattern recognition, and self-organization, into monitoring networks. In the immune-inspiredmonitoring paradigm, a group of autonomous mobile monitoring agents mimic immune cells(such as B-cells) in the natural immune system, interact locally with monitoring environment,and respond to emerging problems through simulated immune responses.