Patrick J. Martin, Ph.D. profile photo

Patrick J. Martin, Ph.D.

Assistant Professor, Department of Electrical and Computer Engineering | Electrical and Computer Engineering

Richmond, VA, UNITED STATES, Engineering West Hall, Room 414

(804) 828-0049

Dr. Martin's research applies control theory and artificial intelligence to enable the safe deployment of distributed autonomous systems.


Industry Expertise

  • Research
  • Defense

Areas of Expertise

Robotics and AutomationAutonomyMachine learning and Artificial Intelligence for Autonomous SystemsCyber Physical Systems


Georgia Institute of Technology

Ph.D., Electrical and Computer Engineering


University of Maryland

M.S., Electrical Engineering


Hampden-Sydney College

B.S., Physics and Applied Mathematics



  • Institute of Electrical and Electronics Engineers (IEEE)
  • Association for the Advancement of Artificial Intelligence (AAAI)
  • American Society for Engineering Education (ASEE)


Secure computational workflows



view more

Research Grants

Heterogeneous Runtime Monitoring for Detection and Assessment of Emerging Hazards for Increasingly Autonomous Urban Flight Vehicles


Urban Air Mobility (UAM) for passenger carrying operations promises to revolutionize urban transportation by potentially reducing congestion, enabling adaptation to changing transportation needs, and creating new market opportunities. The realization of trustworthy, safe and secure UAM operations will depend on three system-level functions: Monitoring, Assessment, and Mitigation. Our approach synthesizes model- and data-driven techniques to characterize real-time data streams as well as historical trends. We leverage Runtime Verification techniques using the TeSSla language combined with data driven neural network based monitors. Furthermore, the structure and synthesis of these hazard predictors is informed by hazard analysis techniques, e.g. STPA, to ensure proper causal chains.


EGRE691 - Autonomous Cyber-physical Systems

In the near future, autonomous cyber-physical systems will be deployed into many facets of the global economy: from transportation to advanced manufacturing. This course introduces the architectures and algorithms that enable the development and deployment of autonomous cyber-physical systems. Students will develop systems that perceive and take action in the physical world using control-theoretic and machine learning techniques. This course reinforces theory with practical assignments that leverage software tools common in the autonomous system and machine learning communities. At the completion of this course, the student will be able to: - Explain and employ autonomous system architectures - Explain and implement deductive and inductive techniques for sensor processing, control, and decision making - Select, evaluate, and critique advanced concepts from the autonomous systems research literature - Design and develop an autonomous CPS prototype

EGRE254 - Digital Logic Design

An introduction to digital logic design with an emphasis on practical design techniques and circuit implementations. Topics include number representation in digital computers, Boolean algebra, theory of logic functions, mapping techniques and function minimization, design of combinational, clocked sequential and interactive digital circuits such as comparators, counters, pattern detectors, adders and subtractors. Asynchronous sequential circuit concepts are introduced. Students will use the above basic skills in the laboratory to design and fabricate digital logic circuits.

Selected Articles

Heterogeneous Multi-Agent Reinforcement Learning for Unknown Environment Mapping | AAAI 2020 Fall Symposium


C. Wakilpoor, P.J. Martin, C. Rebhuhn, and A. Vu

Reinforcement learning in heterogeneous multi-agent scenarios is important for real-world applications but presents challenges beyond those seen in homogeneous settings and simple benchmarks. In this work, we present an actor-critic algorithm that allows a team of heterogeneous agents to learn decentralized control policies for covering an unknown environment. This task is of interest to national security and emergency response organizations that would like to enhance situational awareness in hazardous areas by deploying teams of unmanned aerial vehicles. To solve this multi-agent coverage path planning problem in unknown environments, we augment a multi-agent actor-critic architecture with a new state encoding structure and triplet learning loss to support heterogeneous agent learning. We developed a simulation environment that includes real-world environmental factors such as turbulence, delayed communication, and agent loss, to train teams of agents as well as probe their robustness and flexibility to such disturbances.

PhysiCloud: A cloud-computing framework for programming cyber-physical systems | IEEE Conference on Control Applications (CCA)


P. Glotfelter, T. Eichelberger, and Patrick J. Martin

This paper presents a cloud-computing inspired framework that facilitates the programming of a deployed cyber-physical system. This framework, PhysiCloud, uses a novel combination of abstractions that hide the implementation details of the underlying cyber-physical system. Additionally, the framework is designed to operate on low-power, mobile systems with resiliency to network failures. Using this system, a controls application developer can focus on their algorithm development and its information dependencies, rather than issues of low level scheduling and communication.

Hybrid Systems Tools for Compiling Controllers for Cyber-Physical Systems | Journal of Discrete Event Dynamical Systems


P. Martin and M. Egerstedt

In this paper, we consider the problem of going from high-level specificationsof complex control tasks for cyber-physical systems to their actual implementation and execution on physical devices. This transition between abstraction levels inevitably results in a specification-to-execution gap, and we discuss two sources for this gap; namely model based and constraint based. For both of these two types of sources,we show how hybrid control techniques provide the tools needed to compile high-level control programs in such a way that the specification-to-execution gap is removed. The solutions involve introducing new control modes into nominal strings of control modes as well as adjusting the control modes themselves.