Pei Zhang

Associate Research Professor

Electrical and Computer Engineering Department

Information Networking Institute

Carnegie Mellon University

peizhang@cmu.edu

Research Interest

My research focuses on cyber-physical systems learning by integrating data, physical knowledge, and hardware systems, while informed by deployments in real-world applications.

Machine learning has become a useful tool for many data-rich problems. However, its use in cyber-physical systems (CPS) has been severely limited because of its need for large amounts of well-labeled data, often tailored to each deployment scenario. While especially challenging for high-dimensional data, the situation is further exacerbated by the complexity and variability of the physical systems being studied and modeled. For example, smart city applications often require significant data to obtain the required robustness for operations in different weather, time of day, users, and cities, etc.

My research enables data science in real physical systems by reducing reliance on initial labeled data through the integration of physical knowledge, the actuation of sensing systems, and the adaptation of data models. My early work, ZebraNet, is considered seminal work in mobile sensor networks for which I received the Test-of-Time award. Currently, my work focus on

  • Fuse physics-based models with empirical models of the system to create more data;
  • Actuate physical sensing hardware to improve data quality for optimal learning;
  • Optimize data adaptation between different application scenarios using the physical understanding of how data distributions change.

The research is informed by real-world applications and deployments using the the structure as sensors, and mobile carriers as sensors.

Carousel imageCarousel imageCarousel imageCarousel image

Structure as Sensors

Structure as Sensors is a new class of sensing systems we enable by our research. We use the structure (e.g., buildings) to acts as the physical sensor element, and use the structural responses (e.g., vibrational movements) to understand the details of physical events (e.g., persons moving in a building, or events around the building). This approach reduces the deployment difficulties of a sensing system but significantly increases the dimensionality of the problem space due to numerous influence factors that change the structural response (e.g., wave propagation in the building, building material, walking speed, occupant activity, operation of machinery, etc.).

Using the structure as a sensor, we have a number of deployments measuring:

  • Track elderly gait for fall prediction
  • Gauge muscular dystrophy in children
  • Pig health monitoring
  • Item identification and association in cashier-less supermarkets
Carousel imageCarousel imageCarousel imageCarousel image

Mobile Carriers as Sensors

Using mobile carrier as Sensors, we utilize the movement of sensors to increase the efficiency of measuring phenomenons that covers a wide area (e.g. cities). Although mobile sensors can allow for this greater coverage, their mobility can also create areas with no coverage and unbalanced coverage for the sensing goals. This requires system coordination and collaboration in order to produce optimal outcomes. As part of this work, we 1) actuate the sensors to optimize sensor placement, movement, and data gathering; and 2) fuse with physical models to generate missing data (e.g. physical model of air flow to model pollution).

Using the mobile carrier as sensors, we developed and deployed a number of applications using a different hardware platforms :

  • Taxi-fleet enabled air pollution monitoring
  • Wild zebra migration tracking
  • Muscular activity and status tracking
  • Micro UAV swarms

Ph.D. Students

Publications and Media Coverage

Click here to for a full list of our

  • Publications (IMWUT, SenSys, IPSN, BuildSys, etc.) and
  • Media coverage of our work (Discovery Channel, CNN, NBC, CBS, Popular Science, etc. )

Select Services

Editorial Board:

  • Associate Editor:
    • Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT);
    • ACM Transaction on Sensor Networks (ToSN).
  • Topic Editor:
    • Frontiers in Built Environment: Structural Sensing: Context-Aware Structures: Understanding the User and Surrounding Situations;
  • Guest Editor:
    • Special issue on "Systems for Smart and Efficient Built Environments": Transaction on Sensor Networks

TPC/GC Chair: General Chairs (SenSys 2018, IPSN 2018, MobiQuitous 2015), Technical Program Committee Chairs (BuildSys 2019, IPSN 2017, MobiCASE 2016, SmartGridComm 2013)

Select Program Committee: SenSys (2018, 2017, 2016), IPSN(2018, 2016, 2015, 2014, 2013), ExtremeCom 2014, Percom 2011, ICCCN 2011.

Teaching

18-843S/14-841: Mobile and Pervasive Computing: This is a course exploring research issues in the newly emerging field of mobile computing. Many traditional areas of computer science and computer engineering are impacted by the constraints and demands of mobility. Examples include network protocols, power management, user interfaces, file access, ergonomics, and security. This will be an advanced course in the truest sense --- most, if not all, the topics discussed will be ones where there is little consensus in the research community on the best approaches. The course will also offer significant hand-on experience in this area. Each student will have to present and lead the discussion on a number of papers. Students will work in groups of three under the guidance of a mentor on a hands-on project. Each student will also be required to write one of two documents: (a) a research proposal (similar in spirit to an NSF proposal) on an idea in mobile computing or (b) a short business plan for a commercial opportunity in mobile computing. Grading will be based on the quality of the presentations, the project, and the proposal or business plan.

18-644/14-840 : Mobile Hardware for Software Engineers: This course covers applications of mobile hardware systems and the hardware associated with these systems. The course enables students 1) to analyze the implications of mobile hardware capabilities and restrictions in order to plan and develop mobile applications, 2) to propose and justify new ideas in the mobile space, and 3) to expose students to a range of mobile systems. Students will be able to devise and interface simple hardware additions to enable new applications. The course covers the elements of embedded systems development, such as hardware fundamentals, system development, as well mobile topics such as power management, machine-to-machine communication, and applications. Student teams will undertake small HW/SW interfacing projects on Arduino to sharpen their experience, and shape and build a novel application with the faculty. Unlike a conventional hardware course, the course would instead focus on the system and software implications, rather than the hardware components (i.e. CPU and radio). Prerequisites: Some understanding of basic electrical terminology; Java programming and C programming desired.

Select Awards

  • Best Student Paper Award, ASCE EMI Dynamics Committee (2020)
  • Best 2019 Journal Paper award, SHM/NDE Technical Committee (2020)
  • Best Paper Award, The ACM/IEEE International Conference on Internet of Things Design and Implementation (IoTDI) (2020)
  • Best Student Paper Award, ASCE EMI Dynamics Committee (2019)
  • Best Paper Award, IEEE International Conference on Machine Learning and Applications (ICMLA 2018)
  • Best Student Paper Award, ASCE EMI Dynamics Committee (2018)
  • Test-of-Time Award, SenSys 2017
  • Audience Choice Award, BuildSys 2017
  • Distinguished Service Award, IPSN 2018
  • Google Faculty Award 2013, 2016
  • NSF CAREER Award 2012