Research

Overview

My research interests include control theory and formal methods, with application in verification, falsification, and control synthesis for cyber-physical systems. I am particularly interested in developing algorithms that design and analyze safety-critical systems with imperfect sensor models (e.g., delayed event detection, noisy measurements, missing data, vision-based sensing).

Control Synthesis & Falsification under Imperfect Information

• We are working on the problem of synthesizing correct-by-construction controllers for systems with noisy, lossy, and event-triggered state information. Our goal is to develop computationally efficient algorithms to design controllers that make decisions with imperfect information yet still provably guarantee the satisfaction of the specification.

1. L. Yang and N. Ozay, “Safety control synthesis for systems with missing measurements”, 7th IFAC Conference on Analysis and Design of Hybrid Systems (ADHS), Brussels, Belgium, July 2021.

2. L. Yang and N. Ozay, “Efficient safety control synthesis with imperfect state information”, Proc. 59th IEEE Conference onDecision and Control (CDC), Jeju Island, South Korea, December 2020.

3. L. Yang, X. Cui, A. Avestruz, and N. Ozay, “Correct-by-construction control synthesis for buck converters with even-triggered state measurement”, Proc. American Control Conference (ACC), Philadelphia, PA, July 2019.

• We also study the problem of falsifying cyber-physical systems against safety requirements in the imperfect information setting. We draw ideas from control theory, formal methods and optimization and develop computational tools to analyze safety-critical systems’ vulnerability and to efficiently find nontrivial corner cases, where a safety violation is likely but difficult to reveal via Monte-Carlo simulation. Our approach applies to systems with {optimization, learning, vision}-based components in the loop, which are hard to formally verify.

1. G. Chou*, Y. E. Sahin*, L. Yang*, K. J. Rutledge, P. Nilsson, and N. Ozay, “Using control synthesis to generate corner cases: A case study on autonomous driving”, ACM SIGBED International Conference on Embedded Software (EMSOFT), Torino, Italy, October 2018. *Equal contribution.

2. L. Yang and N. Ozay, “Synthesis-guided adversarial scenario generation for gray-box feedback control systems with sensing imperfections”, ACM Transactions on Embedded Computing Systems (ESWEEK-TECS special issue), 20(5s): Article 102, October 2021.

3. L. Yang, H. Zhang, J. B. Jeannin and N. Ozay, “Efficient backward reachability using the minkowski difference of constrained zonotopes”, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (ESWEEK-TCAD special issue), 2022. to appear.

4. L. Yang and N. Ozay, “Scalable zonotopic under-approximation of backward reachable sets for uncertain linear systems”, IEEE Control Systems Letters, 6: 1555-1560, October 2021 (arXiv version).

Correct-By-Construction Fault-tolerant Control & Fault Detection with Linear Temporal Logic Constraints

We study the problem of fault-tolerant control under delayed fault detection. Our goal is to achieve a gracefully degraded system that performs complex tasks formally specified by linear temporal logic. We explore the structural properties in thespecifications and the system dynamics to develop efficient algorithms that i) provideguarantees on finite detectability of faults, and ii) synthesize fault-tolerantcontrollers for hybrid systems using abstraction-based techniques and the supervisory control theory.

1. R. Niu, S. M. Hassaan, L. Yang, Z. Jin and S. Z. Yong, “Model Discrimination of Switched Nonlinear Systems with Temporal Logic-Constrained Switching”, IEEE Control System Letters, January 2021.

2. L. Yang and N. Ozay, “Fault detectability analysis of switched affine systems with linear temporal logic constraints”, Proc. 58th IEEE Conference on Decision and Control (CDC), Nice, France, December 2019.
   

3. L. Yang and N. Ozay, “Provably-correct fault-tolerant control with delayed information”, Proc. 56th IEEE Conference on Decision and Control (CDC), Melbourne, Australia, December 2017.

4. L. Yang and N. Ozay, “Fault-tolerant output-feedback path planning with temporal logic constraints”, Proc. 57th IEEE Conference on Decision and Control (CDC), Miami Beach, FL, December 2018.
   

5. L. Yang and N. Ozay, “Tight decomposition functions for mixed monotonicity”, Proc. 58th IEEE Conference on Decision and Control (CDC), Nice, France, December 2019.

6. L. Yang, O. Mickelin, and N. Ozay, “On sufficient conditions for mixed monotonicity”, IEEE Transactions on Automatic Control, 64(12): 5080-5085, December 2019.

7. L. Yang, A. Karnik, B. Pence, M. T. B. Waez, and N. Ozay, “Fuel cell thermal management: Modeling, specifications and correct-by-construction control synthesis”, IEEE Transactions on Control Systems Technology, 28(5): 1638-1651, September 2020.

8. L. Yang, N. Ozay, and A. Karnik, “Synthesis of fault-tolerant switching protocols for vehicle engine thermal management”, Proc. American Control Conference (ACC), Boston, MA, July 2016.