Michael Bauer

Professor Bauer’s research interests include:

1. The design, implementation and management of distributed applications and systems, particularly:
   1. Policy-driven autonomic computing for developing self-managing systems and applications;
   2. Applications and systems dealing with the Internet of Things and cloud computing;
   3. Cloud-based computational services for near real-time computing of results from stream-based data sources, such as sensors;
   4. Managing, monitoring and modelling smart city environments and applications.
2. The development of methods to enhance intelligent driver assistance systems, including:
   1. Automated methods for inferring and predicting driving behaviour;
   2. The use of vehicle-to-vehicle and vehicle-to-infrastructure communication for data sharing among vehicles in order to enhance driver assistance systems.
3. The design, implementation and use of data analytics methods, including machine learning, for the analysis of health and medical data in collaboration with researchers from health and medical sciences, including:
   1. Use of unsupervised and supervised methods for analysis of data;
   2. Use of machine learning methods for discovering associations and developing predictive models;
   3. Use of mobile applications for the collection of data about students with the aim of exploring indicators of changes in student behavior that may be suggestive of mental stress.

Recent Journal Publications

1. Khairdoost, M. Shirpour, M. Bauer, S. Beauchemin. Real-Time Maneuver Prediction Using LSTM. IEEE Transactions on Intelligent Vehicles, 2020, (online early access).
2. T Prado and M. Bauer. ARPS: A Framework for Development, Simulation, Evaluation, and Deployment of Multi-Agent Systems, Applied. Science (special issue on Multi-agent Systems). 2019, 9(21), https://doi.org/10.3390/app9214483, open access.
3. Lisa L. Van Loon, N. Stewart McIntyre, Michael Bauer, Nathaniel A. Sherry, Neil R. Banerjee. Peakaboo: Advanced software for the interpretation of X-ray fluorescence spectra from synchrotrons and other intense X-ray sources. Software Impacts, Elsevier, Vol. 2.,2019, doi: https://doi.org/10.1016/j.simpa.2019.100010.
4. K. Nicholson, A. Terry, M. Fortin, T. Williamson, M. Bauer, A. Thind. Prevalence, Characteristics, and Patterns of Patients with Multimorbidity: A Retrospective Cohort Analysis in Canada.  British Journal of General Practice (2017 IF: 3.261), 2019, July, 2019, DOI: https://doi.org/10.3399/bjgp19X704657, 10 pages.
5. Zardosht, S. Beauchemin, M. Bauer. Identifying Individual Driver Behavior in Pre-Turning Maneuvers Using In-Vehicle CANbus Signals. Journal of Advanced Transportation, https://doi.org/10.1155/2018/5020648,Wiley, November, 2018, 10 pages.
6. Tighe, M. Bauer. Topology and Application Aware Dynamic VM Management in the Cloud.  Journal of Grid Computing, Springer, doi=10.1007/s10723-017-9397-z, 2017, pp. 1-22; (2016 Impact Factor: 2.766).

Recent Refereed Conference Publications

1. Shirpour, S. Beauchemin, M. Bauer. What Does Visual Gaze Attend to During Driving? 7th Int. Conf. on Vehicular Technology and Intelligent Transport Systems (VEHITS), 2021, to appear.
2. Lopes, V. Stroele, R. Braga, J. David, M. Bauer. Identifying and Recommending Experts Using a Syntactic-Semantic Analysis Approach. IEEE Computer Supported Cooperative Work in Design (IEEE CSCWD), IEEE, 2021, to appear.
3. Shirpour, S. Beauchemin, M. Bauer. Road Lane Detection and Classification in Urban and Suburban Areas Based on CNNs, 16th Int. Joint Conf. on Computer Vision, Imaging, and Computer Graphics Theory and Applications (VISAPP), 2021, to appear.
4. Khairdoost, S. Beauchemin, M. Bauer. A Probabilistic Model for Visual Driver Gaze Approximation from Head Pose Estimation. IEEE Connected and Automated Vehicles Symposium, IEEE, 2020, to appear.
5. Alfar, Y. Kotb, M. Bauer, Smart City Sensor Network Control and Optimization Using Intelligent Agents, Future of Information and Communications Conference (FICC) 2021, April, 2021, to appear.
6. Khatouni, M. Bauer, H. Lutfiyya, Indoor Temperature Characterization and its Implication on Power Consumption in a Campus Building, 7^th International Conference on Internet of Things: Systems, Management and Security (IOTSMS 2020), December, 2020, .pp. 6-13
7. Elsakhawy and M. Bauer. FaaS2F: A Framework for Defining Execution-SLA is Serverless Computing. IEEE Cloud Summit 2020, IEEE, 2020, October, to appear.

Courses taught in 2020-21:

• CS1026: Computer Science Fundamentals I

•  CS1026: Computer Science Fundamentals I (online)

Automated Management of Smart City Infrastructure.  This project looks to develop methods, preferably automatic methods, for the efficiently managing smart city infrastructure: the sensors, actuators, networks, computers, middleware, services and applications.  This includes development of algorithms for efficient processing, resource management, use of policy-based autonomic methods, and machine learning for prediction faults or problems.  The aim is to have methods and prototypes that can be deployed to help cities manage this kind of complex IoT, computing and application environment.  Algorithms and methods are evaluated through the development of simulators that simulate aspects of a smart city, including the sensor/actuator infrastructure (IoT), computers, networks, middleware and applications.

Augmenting Advanced Driver Assistance Systems Through Driver Models.  This project currently focuses on building models of driving behavior that try to capture driver intent through driver gaze.  Models are built based on vehicle data (CANBUS), stereo imagery of the driving environment and driver gaze.  Current work focuses on the use of machine learning methods to predict driver maneuvers.  Models are developed off-line using data collected from actual driving sequences.   Future work will focus on evaluating these models within actual vehicles and the development of models to not only predict driver maneuvers but to assess any potential risk or danger from those maneuvers.  This work is done with Dr. Steven Beauchemin and we are currently collaborating with Dr. Taufiq Rahman of the National Research Council.

Smart Healthy Campus.  In this project we are interested in identifying data from a mobile phone that can help identify students that may be under stress or anxious.  Mobile applications have been developed that deliver periodic questions to students about their feelings.  Along with the responses to the questions, a variety of data about the student’s phone, usage, general activity are collected.  The aim is to look for association between changes in the way the phone is used with student feelings.  This work is done in collaboration with researchers from Health Sciences.

1. Peakaboo. Lisa L. Van Loon, N. Stewart McIntyre, Michael Bauer, Nathaniel A. Sherry, Neil R. Banerjee. Peakaboo: Advanced software for the interpretation of X-ray fluorescence spectra from synchrotrons and other intense X-ray sources. Software Impacts, Elsevier, Vol. 2.,2019, doi: https://doi.org/10.1016/j.simpa.2019.100010.
2. Multimorbidity Cluster Analysis Toolkit. http://www.csd.uwo.ca/faculty/bauer/Multimorbidity.htm.  Last accessed July, 2019.  Described in: Nicholson, A. Thind, A. Terry, M. Fortin, T. Williamson, M. Bauer.  The Multimorbidity Cluster Analysis Tool: Identifying Combinations and Permutations of Multiple Chronic Diseases Using a Record-Level Computational Analysis.  Journal of Innovation in Health Informatics, 2017, doi=10.14236/jhi.v24i4.962, Vol. 24 , No. 4, pp. 339-343.
3. DCSim. Distributed and Grid Systems (DiGS) Research Group, Western University. [Online]. Not available. Described in: G. Keller, M. Tighe, H. Lutfiyya and M. Bauer, DCSim: A data centre simulation tool, 2013 IFIP/IEEE International Symposium on Integrated Network Management (IM 2013), 2013, pp. 1090-1091.