“Composition and Decomposition for Verified Autonomy”

Dr. Kevin Leahy

Assistant Professor, Robotics Engineering

Worcester Polytechnic Institute

When: 10:50 AM Friday, March 13, 2026

Where: EB Event Space — Zoom Link

(Milk and Cookies available for attendees)

Abstract:

Machine learning failures increasingly compromise autonomous systems, yet the black box nature of learned components makes traditional verification approaches inadequate. This challenge is particularly acute in safety -critical applications like autonomous vehicles and robotic systems, where unverified ML components can lead to catastrophic failures despite high average-case performance. We address this verification gap through compositional system design, treating learned components as modules within larger, formally analyzable architectures. Unlike end-to-end learning approaches that are opaque and difficult to verify, compositional design enables modular reasoning: each component can be verified independently with guaranteed properties that compose into system- level safety assurances. This modularity makes verification tractable while preserving the flexibility of machine learning. This talk presents two complementary approaches for verifiable autonomous systems. First, I demonstrate bottom-up composition in reinforcement learning, where we synthesize complex behaviors by formally composing verified sub-policies with provable guarantees on the resulting system behavior. Second, I present top-down decomposition for perception-based control, separating vision systems from control modules to enable independent verification of each component while maintaining end-to-end performance guarantees.

Biography:

Kevin Leahy is an Assistant Professor in Robotics Engineering at Worcester Polytechnic Institute, focusing involves AI for autonomous systems, with an emphasis on formal methods and multi-agent systems. He received his Ph.D. degree in Mechanical Engineering in 2017 from Boston University. From 2017 to 2023, he was a member of the Technical Staff at MIT Lincoln Laboratory, working on learning decentralized control strategies for multi-agent systems, researching collision avoidance in aviation, and planning for heterogeneous teams from high-level specifications. He currently serves as Associate Co -chair on the IEEE RAS Technical Committee for Verification of Autonomous Systems. He is a recipient of the 2025 ONR YIP award.

“Composition and Decomposition for Verified Autonomy”

Dr. Kevin Leahy

Assistant Professor, Robotics Engineering

Worcester Polytechnic Institute

When: 10:50 AM Friday, March 13, 2026

Where: EB Event Space — Zoom Link

(Milk and Cookies available for attendees)

Abstract:

Machine learning failures increasingly compromise autonomous systems, yet the black box nature of learned components makes traditional verification approaches inadequate. This challenge is particularly acute in safety -critical applications like autonomous vehicles and robotic systems, where unverified ML components can lead to catastrophic failures despite high average-case performance. We address this verification gap through compositional system design, treating learned components as modules within larger, formally analyzable architectures. Unlike end-to-end learning approaches that are opaque and difficult to verify, compositional design enables modular reasoning: each component can be verified independently with guaranteed properties that compose into system- level safety assurances. This modularity makes verification tractable while preserving the flexibility of machine learning. This talk presents two complementary approaches for verifiable autonomous systems. First, I demonstrate bottom-up composition in reinforcement learning, where we synthesize complex behaviors by formally composing verified sub-policies with provable guarantees on the resulting system behavior. Second, I present top-down decomposition for perception-based control, separating vision systems from control modules to enable independent verification of each component while maintaining end-to-end performance guarantees.

Biography:

Kevin Leahy is an Assistant Professor in Robotics Engineering at Worcester Polytechnic Institute, focusing involves AI for autonomous systems, with an emphasis on formal methods and multi-agent systems. He received his Ph.D. degree in Mechanical Engineering in 2017 from Boston University. From 2017 to 2023, he was a member of the Technical Staff at MIT Lincoln Laboratory, working on learning decentralized control strategies for multi-agent systems, researching collision avoidance in aviation, and planning for heterogeneous teams from high-level specifications. He currently serves as Associate Co -chair on the IEEE RAS Technical Committee for Verification of Autonomous Systems. He is a recipient of the 2025 ONR YIP award.

“Combining Human Demonstrations and Interventions for Safe Training of Autonomous Systems in Real-Time”

Dr. John Valasek

Professor of Aerospace Engineering, Director - Vehicle Systems & Control Laboratory

Texas A&M University

When: 10:50 AM Friday, February 20, 2026

Where: EB Event Space — Zoom Link

(Milk and Cookies available for attendees)

Abstract:

Reinforcement Learning (RL) has yielded many recent successes in solving complex tasks that meet and exceed the capabilities of human counterparts, demonstrated in video game environments (Mnih et al. 2015), robotic manipulators (Andrychowicz et al. 2018), and various open-source simulated scenarios (Lillicrap et al. 2015). However, these RL approaches are sample inefficient and slow to converge to this impressive behavior, limited significantly by the need to explore potential strategies through trial and error, which produces initial performance significantly worse than human counterparts. The resultant behavior that is initially random and slow to reach proficiency is poorly suited to various situations, such as physically embodied ground and air vehicles or in scenarios where sufficient capability must be achieved in short time spans. In such situations, the random exploration of the state space of an untrained agent can result in unsafe behaviors and catastrophic failure of a physical system, potentially resulting in unacceptable damage or downtime.

One approach for overcoming these limitations is to utilize demonstrations of desired behavior from a human data source (or potentially some other agent) to initialize the learning agent to a significantly higher level of performance than is yielded by a randomly initialized agent. Behavior Cloning (BC) learns a mapping between the state-action pairs contained in the set of demonstrations to mimic the behavior of the demonstrator but is limited by the quality and quantity of the demonstrations provided and is only improved by providing additional, high-quality demonstrations.

This presentation introduces the Cycle-of-Learning (CoL) framework that uses an actor-critic architecture with a loss function that combines behavior cloning and 1-step Q- learning losses with an off-policy pre-training step from human demonstrations. This enables transition from behavior cloning to reinforcement learning without performance degradation and improves reinforcement learning in terms of overall performance and training time. Additionally, we carefully study the composition of these combined losses and their impact on overall policy learning. We show that our approach outperforms state- of-the-art techniques for combining behavior cloning and reinforcement learning for both dense and sparse reward scenarios. Our results also suggest that directly including the behavior cloning loss on demonstration data helps to ensure stable learning and ground future policy updates.

Biography:

John Valasek is Director, Vehicle Systems & Control Laboratory (https://vscl.tamu.edu), Thaman Professor of Undergraduate Teaching Excellence, Professor of Aerospace Engineering, and member of the Honors Faculty at Texas A&M University (TAMU). He has been conducting autonomy and flight controls research of Manned and Unmanned Air Systems (UAS) in both Industry and Academia for 41 years, first as a Flight Control Engineer for the Northrop Corporation, Aircraft Division. His research is focused on bridging the scientific gaps between traditional computer science topics and aerospace engineering topics, while achieving a high degree of closure between theory and experiment. John is a Fellow of AIAA and is serving as Deputy Director of the Information Systems Group. John earned the B.S. degree in Aerospace Engineering from California State Polytechnic University, Pomona in 1986 and the M.S. degree with honors and the Ph.D. in Aerospace Engineering from the University of Kansas, in 1991 and 1995 respectively.

“Combining Human Demonstrations and Interventions for Safe Training of Autonomous Systems in Real-Time”

Dr. John Valasek

Professor of Aerospace Engineering, Director - Vehicle Systems & Control Laboratory

Texas A&M University

When: 10:50 AM Friday, February 20, 2026

Where: EB Event Space — Zoom Link

(Milk and Cookies available for attendees)

Abstract:

Reinforcement Learning (RL) has yielded many recent successes in solving complex tasks that meet and exceed the capabilities of human counterparts, demonstrated in video game environments (Mnih et al. 2015), robotic manipulators (Andrychowicz et al. 2018), and various open-source simulated scenarios (Lillicrap et al. 2015). However, these RL approaches are sample inefficient and slow to converge to this impressive behavior, limited significantly by the need to explore potential strategies through trial and error, which produces initial performance significantly worse than human counterparts. The resultant behavior that is initially random and slow to reach proficiency is poorly suited to various situations, such as physically embodied ground and air vehicles or in scenarios where sufficient capability must be achieved in short time spans. In such situations, the random exploration of the state space of an untrained agent can result in unsafe behaviors and catastrophic failure of a physical system, potentially resulting in unacceptable damage or downtime.

One approach for overcoming these limitations is to utilize demonstrations of desired behavior from a human data source (or potentially some other agent) to initialize the learning agent to a significantly higher level of performance than is yielded by a randomly initialized agent. Behavior Cloning (BC) learns a mapping between the state-action pairs contained in the set of demonstrations to mimic the behavior of the demonstrator but is limited by the quality and quantity of the demonstrations provided and is only improved by providing additional, high-quality demonstrations.

This presentation introduces the Cycle-of-Learning (CoL) framework that uses an actor-critic architecture with a loss function that combines behavior cloning and 1-step Q- learning losses with an off-policy pre-training step from human demonstrations. This enables transition from behavior cloning to reinforcement learning without performance degradation and improves reinforcement learning in terms of overall performance and training time. Additionally, we carefully study the composition of these combined losses and their impact on overall policy learning. We show that our approach outperforms state- of-the-art techniques for combining behavior cloning and reinforcement learning for both dense and sparse reward scenarios. Our results also suggest that directly including the behavior cloning loss on demonstration data helps to ensure stable learning and ground future policy updates.

Biography:

John Valasek is Director, Vehicle Systems & Control Laboratory (https://vscl.tamu.edu), Thaman Professor of Undergraduate Teaching Excellence, Professor of Aerospace Engineering, and member of the Honors Faculty at Texas A&M University (TAMU). He has been conducting autonomy and flight controls research of Manned and Unmanned Air Systems (UAS) in both Industry and Academia for 41 years, first as a Flight Control Engineer for the Northrop Corporation, Aircraft Division. His research is focused on bridging the scientific gaps between traditional computer science topics and aerospace engineering topics, while achieving a high degree of closure between theory and experiment. John is a Fellow of AIAA and is serving as Deputy Director of the Information Systems Group. John earned the B.S. degree in Aerospace Engineering from California State Polytechnic University, Pomona in 1986 and the M.S. degree with honors and the Ph.D. in Aerospace Engineering from the University of Kansas, in 1991 and 1995 respectively.

“Safe and Stable: A New Framework for Enforcing State Constraints”

Dr. Ankit Goel

Assistant Professor of Mechanical Engineering

University of Maryland

When: 3:50 PM Friday, September 26, 2025

Where: EB Event Space — Zoom Link

(Milk and Cookies available for attendees)

Abstract:

“Over the past decade, Control Barrier Functions (CBFs) have emerged as a leading framework for enforcing state constraints in control system design. The central principle is the enforcement of forward invariance of a “safe set” defined by the constraints. The origins of this idea trace back to Nagumo’s invariance condition in the 1940s, followed by a brief focus on Barrier Lyapunov Functions in the 2000s, and the resurgence of CBFs in the mid-2010s. While CBFs provide a powerful tool for safety enforcement, they operate primarily as filters that modify a nominal control input to satisfy invariance conditions. This separation of safety enforcement from controller design can alter closed-loop stability and, in complex systems, may fail to guarantee safety.

In this talk, I will introduce a new framework for constrained control that unifies stability and safety enforcement within a single design procedure. By integrating state constraints directly into the control synthesis step, the approach preserves the rigor of classical Lyapunov-based analysis while ensuring forward invariance of the prescribed safe set, opening new directions for the reliable control of safety-critical dynamical systems.”

Biography:

Dr. Ankit Goel is an Assistant Professor of Mechanical Engineering at the University of Maryland, Baltimore County. He received his B.E. in Mechanical Engineering from Delhi College of Engineering in 2009, followed by M.S. and Ph.D. degrees in Aerospace Engineering from the University of Michigan, Ann Arbor, in 2014 and 2019, respectively. Dr. Goel joined the UMBC faculty in 2021, where he directs the Estimation, Control, and Learning Laboratory (ECLL). His research explores both model-free, data-driven, learning-based approaches and model-based methods for the design of control systems in complex dynamical environments.

“Safe and Stable: A New Framework for Enforcing State Constraints”

Dr. Ankit Goel

Assistant Professor of Mechanical Engineering

University of Maryland

When: 3:50 PM Friday, September 26, 2025

Where: EB Event Space — Zoom Link

(Milk and Cookies available for attendees)

Abstract:

“Over the past decade, Control Barrier Functions (CBFs) have emerged as a leading framework for enforcing state constraints in control system design. The central principle is the enforcement of forward invariance of a “safe set” defined by the constraints. The origins of this idea trace back to Nagumo’s invariance condition in the 1940s, followed by a brief focus on Barrier Lyapunov Functions in the 2000s, and the resurgence of CBFs in the mid-2010s. While CBFs provide a powerful tool for safety enforcement, they operate primarily as filters that modify a nominal control input to satisfy invariance conditions. This separation of safety enforcement from controller design can alter closed-loop stability and, in complex systems, may fail to guarantee safety.

In this talk, I will introduce a new framework for constrained control that unifies stability and safety enforcement within a single design procedure. By integrating state constraints directly into the control synthesis step, the approach preserves the rigor of classical Lyapunov-based analysis while ensuring forward invariance of the prescribed safe set, opening new directions for the reliable control of safety-critical dynamical systems.”

Biography:

Dr. Ankit Goel is an Assistant Professor of Mechanical Engineering at the University of Maryland, Baltimore County. He received his B.E. in Mechanical Engineering from Delhi College of Engineering in 2009, followed by M.S. and Ph.D. degrees in Aerospace Engineering from the University of Michigan, Ann Arbor, in 2014 and 2019, respectively. Dr. Goel joined the UMBC faculty in 2021, where he directs the Estimation, Control, and Learning Laboratory (ECLL). His research explores both model-free, data-driven, learning-based approaches and model-based methods for the design of control systems in complex dynamical environments

“Soft, Shape-Changing Truss Robots”

Dr. Nathan Usevitch

BYU Department of Mechanical Engineering

When: 3:50 PM Friday, April 4, 2025

Where: EB Event Space -- Zoom Link

(Milk and Cookies available for attendees)

Abstract:

Should robots mimic humans, or can we design entirely new robotic forms? While humanoid robots excel in human spaces, engineering advances allow us to build radically different machines for challenging environments. In this talk, Dr. Usevitch will explore the design of truss-like inflatable robots for lunar operations. These robots can deploy from a compact volume, adapt their shape, and offer a high degree of controllability. These robots could one day form modular, reconfigurable robotic structures that perform a variety of tasks supporting human presence on the moon. Dr. Usevitch will discuss how computational modeling and mechanical design combine to push the boundaries of robotic capability.

Biography:

Nathan Usevitch joined the faculty of the BYU Mechanical Engineering Department in 2023. His research focuses on designing new types of robotic systems, with a focus on robots made from compliant materials and robots that directly interact with humans. Before BYU, Dr. Usevitch worked for Meta Reality Labs designing haptic gloves for virtual reality. He received a Master’s (2017) and PhD (2020) in Mechanical Engineering from Stanford University. His PhD thesis was on soft, shape changing robots. He received his bachelor's degree in mechanical engineering from BYU.

“Soft, Shape-Changing Truss Robots”

Dr. Nathan Usevitch

BYU Department of Mechanical Engineering

When: 3:50 PM Friday, April 4, 2025

Where: EB Event Space -- Zoom Link

(Milk and Cookies available for attendees)

Abstract:

Should robots mimic humans, or can we design entirely new robotic forms? While humanoid robots excel in human spaces, engineering advances allow us to build radically different machines for challenging environments. In this talk, Dr. Usevitch will explore the design of truss-like inflatable robots for lunar operations. These robots can deploy from a compact volume, adapt their shape, and offer a high degree of controllability. These robots could one day form modular, reconfigurable robotic structures that perform a variety of tasks supporting human presence on the moon. Dr. Usevitch will discuss how computational modeling and mechanical design combine to push the boundaries of robotic capability.

Biography:

Nathan Usevitch joined the faculty of the BYU Mechanical Engineering Department in 2023. His research focuses on designing new types of robotic systems, with a focus on robots made from compliant materials and robots that directly interact with humans. Before BYU, Dr. Usevitch worked for Meta Reality Labs designing haptic gloves for virtual reality. He received a Master’s (2017) and PhD (2020) in Mechanical Engineering from Stanford University. His PhD thesis was on soft, shape changing robots. He received his bachelor's degree in mechanical engineering from BYU.

“Enabling Advanced Air Mobility for Uncrewed Platforms”

Dr. Anusna Chakraborty - Senior Research and Development Scientist

UtopiaCompression Corporation

When: 3:50 PM Friday, March 28, 2025

Where: CB 256 -- Zoom Link

(Milk and Cookies available for attendees)

Abstract:

The rapid expansion of Unmanned Aerial Vehicle (UAV) applications has accelerated in recent years. This growth is primarily driven by the increasing availability of diverse algorithms, both classical and machine learning-based, which enable real-time decision-making through miniaturized sensors. However, several practical challenges must be overcome to ensure the safe and efficient large-scale deployment and integration of these vehicles into the National Airspace. In this talk, Dr. Chakraborty will explore key enabling technologies and their role in addressing these challenges.

Biography:

Dr. Anusna Chakraborty is a Senior Research and Development Scientist at UtopiaCompression Corporation, a research-based company working on different projects involving computer vision, machine learning, autonomous systems, and bio-medical devices with the U.S. Government. She received her Ph.D. in Aerospace Engineering from the University of Cincinnati in 2020. Her primary focus during graduate studies was on the development of multi-agent cooperative localization and navigation in harsh and contested environments. She completed her Bachelor in Technology from St. Thomas’ College of Engineering and Technology in Kolkata, India in 2012. Currently, she works on several different projects involving autonomous vehicle operations, estimation, path-planning, and navigation. She has authored several scientific publications in the domain of cooperative localization, formation control, and GPS-denied navigation.

“Enabling Advanced Air Mobility for Uncrewed Platforms”

Dr. Anusna Chakraborty - Senior Research and Development Scientist

UtopiaCompression Corporation

When: 3:50 PM Friday, March 28, 2025

Where: CB 256 -- Zoom Link

(Milk and Cookies available for attendees)

Abstract:

The rapid expansion of Unmanned Aerial Vehicle (UAV) applications has accelerated in recent years. This growth is primarily driven by the increasing availability of diverse algorithms, both classical and machine learning-based, which enable real-time decision-making through miniaturized sensors. However, several practical challenges must be overcome to ensure the safe and efficient large-scale deployment and integration of these vehicles into the National Airspace. In this talk, Dr. Chakraborty will explore key enabling technologies and their role in addressing these challenges.

Biography:

Dr. Anusna Chakraborty is a Senior Research and Development Scientist at UtopiaCompression Corporation, a research-based company working on different projects involving computer vision, machine learning, autonomous systems, and bio-medical devices with the U.S. Government. She received her Ph.D. in Aerospace Engineering from the University of Cincinnati in 2020. Her primary focus during graduate studies was on the development of multi-agent cooperative localization and navigation in harsh and contested environments. She completed her Bachelor in Technology from St. Thomas’ College of Engineering and Technology in Kolkata, India in 2012. Currently, she works on several different projects involving autonomous vehicle operations, estimation, path-planning, and navigation. She has authored several scientific publications in the domain of cooperative localization, formation control, and GPS-denied navigation.

"Advances to Tasking, Coordination, and Safety for Multi-Agent Systems from High-Level Specifications"

Dr. Mitchell Black - Technical Staff, Integrated Missile Defense Technology Group

Paula Donovan - Group Leader, Integrated Missile Defense Technology Group

Dr. Zachary Serlin - Technical Staff, Integrated Missile Defense Technology Group

MIT Lincoln Laboratory

When: 12:50 PM Thursday, February 13, 2025

Where: EB Event Space -- Zoom Link

(Milk and Cookies available for attendees)

Abstract:

"Drive from point A to point B." "Eventually reach the goal and always avoid collisions." Recent advances in formal methods have significantly enhanced the design, coordination, and safety of multi-agent autonomous systems, particularly in the context of satisfying high-level task specifications such as these. This talk explores the formalization, coordination, and enforcement of task-level objectives and safety guarantees from high-level specifications through Capability Temporal Logic, a framework for specifying complex, time-dependent behaviors in dynamic environments, and Control Barrier Functions, which serve as a tool for preserving safety at runtime. Special attention will be given to recent experimental demonstrations of coordinated autonomy for Unmanned Surface Vessels (USVs) conducted by MIT Lincoln Laboratory on the Charles River in Boston.

Biographies:

Mitchell Black is a Technical Staff member at MIT Lincoln Laboratory in the Integrated Missile Defense Technology group, where he works on problems related to safe autonomy. Previously, he received his B.S. from Tufts University in Mechanical Engineering and Astrophysics in 2016, his M.S. and Ph.D. in Aerospace Engineering from the University of Michigan in 2017 and 2023, and was a Research Scientist with the Cyber-Physical Systems group at Toyota North America (2022-2024). In 2016, Mitchell was awarded the NCAA Walter Byers Postgraduate Scholarship for combined achievement in academics and athletics on the Tufts University Track & Field team.

Paula J. Donovan leads the Integrated Missile Defense Technology Group at MIT Lincoln Laboratory. The group's mission is to advance the missile defense of the United States, deployed forces, and allies through analysis, algorithm development, and capability development and testing.

Donovan joined Lincoln Laboratory in 2000, spending her first decade in the Communications Systems Division and her second in the Cyber Security and Information Sciences Division. During this time, she made significant contributions to communication networks and cybersecurity through analysis, metrics development, simulation, algorithm and protocol development, and prototyping. These contributions include modeling the first all-optional network, directing a team in reverse engineering and emulating the Multifunction Advanced Data Link (MADL) waveform, and leading a program to study the workflows of combatant command warfighters and deploy advanced capabilities to them. In 2022, she transitioned to the Air, Missile, and Maritime Defense Technology Division. Donovan now leads research teams in an array of areas advancing missile defense—including through artificial intelligence, autonomy, and cyber technology—and integrating related advances in support of Joint All-Domain Command and Control (JADC2).

Donovan is a Senior Member of IEEE. She holds a patent on an all-optical transceiver for communication networks and has led several teams that received Laboratory recognition awards. She earned her MS in applied mathematics with a concentration in computer science from the University of Massachusetts at Amherst and BA in mathematics with a concentration in pre-medical from the College of the Holy Cross.

Zachary Serlin is a Technical Staff member of the Integrated Missile Defense Technology Group at MIT Lincoln Laboratory. His work has focused on formal guarantees for artificial intelligence-based multi-agent autonomous systems. He has applied novel optimization techniques to Missile Defense Agency battle management problems, and led multiple DARPA and MITLL research and development programs in autonomous systems, one of which was a finalist in the 2022 R&D 100 awards. He holds a BS and MS in mechanical engineering from Tufts University and a PhD in mechanical engineering from Boston University.

"Advances to Tasking, Coordination, and Safety for Multi-Agent Systems from High-Level Specifications"

Dr. Mitchell Black - Technical Staff, Integrated Missile Defense Technology Group

Paula Donovan - Group Leader, Integrated Missile Defense Technology Group

Dr. Zachary Serlin - Technical Staff, Integrated Missile Defense Technology Group

MIT Lincoln Laboratory

When: 12:50 PM Thursday, February 13, 2025

Where: EB Event Space -- Zoom Link

(Milk and Cookies available for attendees)

Abstract:

"Drive from point A to point B." "Eventually reach the goal and always avoid collisions." Recent advances in formal methods have significantly enhanced the design, coordination, and safety of multi-agent autonomous systems, particularly in the context of satisfying high-level task specifications such as these. This talk explores the formalization, coordination, and enforcement of task-level objectives and safety guarantees from high-level specifications through Capability Temporal Logic, a framework for specifying complex, time-dependent behaviors in dynamic environments, and Control Barrier Functions, which serve as a tool for preserving safety at runtime. Special attention will be given to recent experimental demonstrations of coordinated autonomy for Unmanned Surface Vessels (USVs) conducted by MIT Lincoln Laboratory on the Charles River in Boston.

Biographies:

Mitchell Black is a Technical Staff member at MIT Lincoln Laboratory in the Integrated Missile Defense Technology group, where he works on problems related to safe autonomy. Previously, he received his B.S. from Tufts University in Mechanical Engineering and Astrophysics in 2016, his M.S. and Ph.D. in Aerospace Engineering from the University of Michigan in 2017 and 2023, and was a Research Scientist with the Cyber-Physical Systems group at Toyota North America (2022-2024). In 2016, Mitchell was awarded the NCAA Walter Byers Postgraduate Scholarship for combined achievement in academics and athletics on the Tufts University Track & Field team.

Paula J. Donovan leads the Integrated Missile Defense Technology Group at MIT Lincoln Laboratory. The group's mission is to advance the missile defense of the United States, deployed forces, and allies through analysis, algorithm development, and capability development and testing.

Donovan joined Lincoln Laboratory in 2000, spending her first decade in the Communications Systems Division and her second in the Cyber Security and Information Sciences Division. During this time, she made significant contributions to communication networks and cybersecurity through analysis, metrics development, simulation, algorithm and protocol development, and prototyping. These contributions include modeling the first all-optional network, directing a team in reverse engineering and emulating the Multifunction Advanced Data Link (MADL) waveform, and leading a program to study the workflows of combatant command warfighters and deploy advanced capabilities to them. In 2022, she transitioned to the Air, Missile, and Maritime Defense Technology Division. Donovan now leads research teams in an array of areas advancing missile defense—including through artificial intelligence, autonomy, and cyber technology—and integrating related advances in support of Joint All-Domain Command and Control (JADC2).

Donovan is a Senior Member of IEEE. She holds a patent on an all-optical transceiver for communication networks and has led several teams that received Laboratory recognition awards. She earned her MS in applied mathematics with a concentration in computer science from the University of Massachusetts at Amherst and BA in mathematics with a concentration in pre-medical from the College of the Holy Cross.

Zachary Serlin is a Technical Staff member of the Integrated Missile Defense Technology Group at MIT Lincoln Laboratory. His work has focused on formal guarantees for artificial intelligence-based multi-agent autonomous systems. He has applied novel optimization techniques to Missile Defense Agency battle management problems, and led multiple DARPA and MITLL research and development programs in autonomous systems, one of which was a finalist in the 2022 R&D 100 awards. He holds a BS and MS in mechanical engineering from Tufts University and a PhD in mechanical engineering from Boston University.