
NSF Org: |
CMMI Division of Civil, Mechanical, and Manufacturing Innovation |
Recipient: |
|
Initial Amendment Date: | August 11, 2010 |
Latest Amendment Date: | August 11, 2010 |
Award Number: | 1049294 |
Award Instrument: | Standard Grant |
Program Manager: |
George Chiu
CMMI Division of Civil, Mechanical, and Manufacturing Innovation ENG Directorate for Engineering |
Start Date: | August 15, 2010 |
End Date: | July 31, 2013 (Estimated) |
Total Intended Award Amount: | $80,000.00 |
Total Awarded Amount to Date: | $80,000.00 |
Funds Obligated to Date: |
|
History of Investigator: |
|
Recipient Sponsored Research Office: |
1400 TOWNSEND DR HOUGHTON MI US 49931-1200 (906)487-1885 |
Sponsor Congressional District: |
|
Primary Place of Performance: |
1400 TOWNSEND DR HOUGHTON MI US 49931-1200 |
Primary Place of
Performance Congressional District: |
|
Unique Entity Identifier (UEI): |
|
Parent UEI: |
|
NSF Program(s): | CONTROL SYSTEMS |
Primary Program Source: |
|
Program Reference Code(s): |
|
Program Element Code(s): |
|
Award Agency Code: | 4900 |
Fund Agency Code: | 4900 |
Assistance Listing Number(s): | 47.041 |
ABSTRACT
The objective of this EArly-concept Grants for Exploratory Research (EAGER) project is to explore some novel ideas on adaptive sensing and monitoring in wireless sensor networks. The focus will be on formulating monitoring methodology inspired by the natural immune system and agent technology, which could potentially lead to adaptive sensing and monitoring. To mimic adaptive immune response, mathematical models will be investigated to study the collaboration of immune cells and provide guideline for the design of such mechanisms in wireless monitoring networks. The project will look for an agent-based network framework to enable immune-inspired adaptive, self-organizing, and autonomous anomaly detection. Multi-objective optimization algorithms will be studied for the robust control of agent generation and distribution to enhance system responsiveness, anomaly detection probability, and network lifetime. Immune-network-theory-based algorithms will also be investigated for the unsupervised anomaly detection.
The successful completion of this research project will have a great impact on the current ability of adaptive sensing and monitoring. The anticipated adaptive immune response models could potentially allow a monitoring system to adjust the number and type of monitoring agents in response to the changes in monitoring conditions. Multi-objective agent generation and distribution strategies could provide a good balance across responsiveness, anomaly detection probability, and network lifetime. The immune-network-based anomaly detection could offer dynamic updates of a nominal behavior model due to environmental and operational changes. The outcomes of this project will offer clear societal benefits in advancing monitoring technologies for critical infrastructure, power grids, traffic systems, and military applications. The research results will be widely disseminated via the following mechanisms: Web, publication of papers, professional talks, and by incorporating materials into teaching. Educational opportunities for pre-college, undergraduate, and graduate students are integrated throughout the research program. The outreach efforts will impact rural, low income, and pre-college students.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
The primary objective of this project has been to explore novel ideas for adaptive sensing and control of distributed systems, such as structural health monitoring systems and intelligent vehicle networks. Major research activities include the design of an agent-based network framework, mimicking adaptive immune response using agent technology, multi-objective optimization of agent generation and distribution, anomaly detection using pattern recognition, emergent pattern recognition using immune network theory, time series data representation and similarity measures for damage pattern recognition, and adaptive sensing and control of intelligent vehicles.
Excellent progress has been made on all objectivities of the project. Outcomes from research activities are as follows:
- A mobile agent-based wireless monitoring network has been developed by integrating mobile agent network middleware with high computational power sensor units.
- Developed adaptive and autonomous monitoring scheme based on artificial immune system (AIS). In AIS-based monitoring networks, mobile monitoring agents mimic immune cells in the natural immune system and perform monitoring tasks in distributed sensor units by equipped damage detection algorithms. Multi-objective optimization algorithms were developed to control the generation and distribution of mobile monitoring agents.
- Developed unsupervised damage detection algorithm based on artificial immune pattern recognition (AIPR). The AIPR-based damage detection mimics immune recognition mechanisms. The clonal selection principle of the immune system was employed to improve the quality of representative feature vectors.
- Developed novel emergent damage detection algorithm based on immune network theory. The immune-network-based emergent pattern recognition algorithm achieves emergent pattern recognition by dynamically generating an internal image mapping to the input data patterns without the need of specifying the number of patterns in advance.
- Developed pattern recognition algorithms to recognize driving patterns for choosing appropriate control strategies under various environmental and driving conditions. To optimize the energy management of hybrid powertrain, model predictive control algorithm was developed to calculate an optimal sequence of control inputs to the vehicle.
- Developed hardware-in-the-loop simulation system to provide open software architecture for rapid prototyping of hybrid electric vehicle control strategies.
- The research results have been published in 7 journal and 6 conference papers.
- The project provided research opportunities for four PhD, two MS, and two undergraduate students.
Last Modified: 10/21/2013
Modified by: Bo Chen
Please report errors in award information by writing to: awardsearch@nsf.gov.