Hi!, My name is Madhur Behl.
I am an Assistant Professor in the departments of Computer Science, and Systems and Information Engineering, and a member of Link Lab at the University of Virginia.
Before joining UVA, I was a research fellow at Penn and the Co-Founder of Expresso Logic - a NSF SBIR small business building a demand-response recommendation system.
I received my Ph.D. (2015) and M.S. (2012), in Electrical and Systems Engineering, both from the University of Pennsylvania. I recieved my bachelor's degree (2009) in Electronics and Communications Engineering from PEC University of Technology in India.

I am actively recruiting PhD students with a background in machine learning, control systems, optimization, and embedded systems. Opportunites exist to conduct high impact and cross-disciplinary research in Cyber-Physical Systems.


A picture of Madhur Behl

Research

My research develops the foundations of Cyber-Physical Systems (CPS). This involves finding analytical and practical solutions to problems of modeling, control, simulation, operation, safety, and implementation of CPS.
In particular, I’m interested in challenging probelms which involve the use of data-driven methods for CPS. This work is at the confluence of machine learning, control systems, embedded systems, optimization, and statistics.
Applications of my work span from energy-efficient buildings, demand response, and smart cities; to industrial automation, autonomous racing, internet of things, medical devices, and model-based clinical trials.

Data predicitve control (DPC)

Data Predictive Control (DPC)

Bridging the gap between machine learning and predicitve control synthesis.

F1/10 autonomous racing

F1/10 Autonomous Racing

Build, drive, and race 1/10 scale autonomus cars at the limits of control.

Interactive analytics (IAX)

Interactive analytics (IAX)

Answering open-eneded queries using procedural generation and interpretable models.

Data driven modeling for buildings

Model-IQ

Uncertainity propagation from sensor placement to data quality to modeling to control.

MLE+ cosimulation toolbox

MLE+

An open integerated toolbox for cosimulation between MATLAB and EnergyPlus

mod7 chiller plant at UPenn

Green Scheduling

A scalable and lightweight scheduling approach for peak power reduction in buildings.

Enroute the energy router

Enroute

An energy router for peak power reduction in buildings.

mRK RTOS

mRK OS

NanoRK real time operating system on the mbed (ARM) microcontroller

Motionview pvt ltd

Gesture Control

Controlling DICOM (medical imaging) viewers with hand gestures.

News

<
November 2017

I am serving as the associate editor for the SAE International's new journal on Connected and Automated Vehicles

I gave the F1/10 Autonomous Racing tutorials at SenSys 2017, Delft, Netherlands.

Served as the posters and demos chair at BuildSys 2017, Delft, Netherlands.

October 2017

Gave an invited talk on "Data Predictive Control - Bridging Machine and Control in Cyber-Physical Systems" at the 2017 Chesapeake Large-Scale Analytics Conference (CLSAC) . This year's theme was predictive analytics.

Article in The Cavalier Daily on our recent NSF CRISP Award: Multidisciplinary research team explores smart stormwater management

September 2017

I am organizing an invited session on Building Energy Systems at the 2018 American Control Conference.

August 2017

I have received a $2.5M NSF CRISP grant on dMIST: Data-driven Management for Interdependent Stormwater and Transportation Systems, with Jon Goodall, Kamin Whitehouse, Michael Gorman, and Donna Chen

For Fall 2017, I will be teaching a new graduate level special topics course on Prinicples of Modeling for Cyber-Physical Systems.

Invited to attend the 2017 Aspiring NSF Cyber-Physical Systems Principal Investigators' Workshop. [Link]

Attended NSF's New Computer Science Faculty Teaching Workshop held at UC San Diego[Link]. Many thanks to Cynthia Lee, Beth Simon, Leo Porter, and Mark Guzdial.

July 2017

Our manuscript entitled "Data Predictive Control for Cyber-Physical Energy Systems" has been accepted for publication in the ACM Transactions on Cyber-Physical Systems

June 2017

Gave an invited talk on The foundations of Data Predictive Control for Cyber-Physical Systems at the Center for Nonlinear Studies (CNLS) at Los Alamos National Laboratory (LANL) [Link]

My chapter on Data-Driven Modeling, Control, and Tools for Smart Cities has been published in the Smart Cities: Foundations and Pricniples book.

May 2017

Out paper on Data predictive control for building energy consumtpion won the Best Energy Systems Paper Award at the American Control Conference 2017, held in Seattle.

February 2017

I am serving as the poster and demo chair for BuildSys this year.

January 2017

My startup Expresso Logic, has been awarded the Phase 1 NSF Small Business Innovative Research (SBIR) award.

Get in Touch

Contact:

Office: 423 Rice Hall
Phone: +1 (434) 924-1021
Email: madhur.behl@virginia.edu
Walk in office hours:
Mondays from 3-4pm.

Mailing Address:

Madhur Behl
Room 423, Rice Hall,
85 Engineers Way,
University of Virginia
Charlottesville, VA 22903