About

Randy Freeman is a Professor of Electrical and Computer Engineering at Northwestern University, affiliated with the Master of Science in Robotics Program. He holds a Ph.D. in Electrical Engineering from the University of California, Santa Barbara, an M.S. in Electrical Engineering from the University of Illinois at Urbana-Champaign, and a B.S. in Electrical Engineering from Cornell University. His research interests include nonlinear control theory, robust control and optimal control, nonlinear system theory, and control and estimation for multi-agent systems. Freeman has contributed to the field through numerous publications on topics such as distributed swarm formation control, stability analysis of nonlinear feedback systems, and environmental monitoring with distributed robotic agents. His work emphasizes the development of control algorithms and estimation techniques for complex, interconnected systems, impacting areas such as robotics, networked control, and multi-agent coordination.

Research topics

• Computer Science
• Computer network
• Artificial Intelligence
• Theoretical computer science
• Telecommunications
• Discrete mathematics
• Distributed computing
• Human–computer interaction
• Algorithm
• Mathematics

Selected publications

• Cooperative Payload Estimation by a Team of Mocobots

  ArXiv.org · 2025-02-07

  preprintOpen access

  For high-performance autonomous manipulation of a payload by a mobile manipulator team, or for collaborative manipulation with the human, robots should be able to discover where other robots are attached to the payload, as well as the payload's mass and inertial properties. In this paper, we describe a method for the robots to autonomously discover this information. The robots cooperatively manipulate the payload, and the twist, twist derivative, and wrench data at their grasp frames are used to estimate the transformation matrices between the grasp frames, the location of the payload's center of mass, and the payload's inertia matrix. The method is validated experimentally with a team of three mobile cobots, or mocobots.

  PublisherOA PDFDOI

• Cooperative Payload Estimation by a Team of Mocobots

  IEEE Robotics and Automation Letters · 2025-08-11 · 1 citations

  article

  For high-performance autonomous manipulation of a payload by a mobile manipulator team, or for collaborative manipulation with the human, robots should be able to discover where other robots are attached to the payload, as well as the payload's mass and inertial properties. In this paper, we describe a method for the robots to autonomously discover this information. The robots cooperatively manipulate the payload, and the twist, twist derivative, and wrench data at their grasp frames are used to estimate the transformation matrices between the grasp frames, the location of the payload's center of mass, and the payload's inertia matrix. The method is validated experimentally with a team of three mobile cobots, or mocobots.

  PublisherDOI

• Self-Healing Distributed Swarm Formation Control Using Image Moments

  IEEE Robotics and Automation Letters · 2024-05-15 · 2 citations

  article

  Human-swarm interaction is facilitated by a low-dimensional encoding of the swarm formation, independent of the (possibly large) number of robots. We propose using image moments to encode two-dimensional formations of robots. Each robot knows its pose and the desired formation moments, and simultaneously estimates the current moments of the entire swarm while controlling its motion to better achieve the desired group moments. The estimator is a distributed optimization, requiring no centralized processing, and self-healing, meaning that the process is robust to initialization errors, packet drops, and robots being added to or removed from the swarm. Our experimental results with a swarm of 50 robots, suffering nearly 50% packet loss, show that distributed estimation and control of image moments effectively achieves desired swarm formations.

  PublisherDOI

• Self-Healing First-Order Distributed Optimization with Packet Loss

  arXiv (Cornell University) · 2023-08-14

  preprintOpen access

  We describe SH-SVL, a parameterized family of first-order distributed optimization algorithms that enable a network of agents to collaboratively calculate a decision variable that minimizes the sum of cost functions at each agent. These algorithms are self-healing in that their convergence to the correct optimizer can be guaranteed even if they are initialized randomly, agents join or leave the network, or local cost functions change. We also present simulation evidence that our algorithms are self-healing in the case of dropped communication packets. Our algorithms are the first single-Laplacian methods for distributed convex optimization to exhibit all of these characteristics. We achieve self-healing by sacrificing internal stability, a fundamental trade-off for single-Laplacian methods.

  PublisherOA PDFDOI

• Self-Healing Distributed Swarm Formation Control Using Image Moments

  arXiv (Cornell University) · 2023-12-12

  preprintOpen access

  Human-swarm interaction is facilitated by a low-dimensional encoding of the swarm formation, independent of the (possibly large) number of robots. We propose using image moments to encode two-dimensional formations of robots. Each robot knows its pose and the desired formation moments, and simultaneously estimates the current moments of the entire swarm while controlling its motion to better achieve the desired group moments. The estimator is a distributed optimization, requiring no centralized processing, and self-healing, meaning that the process is robust to initialization errors, packet drops, and robots being added to or removed from the swarm. Our experimental results with a swarm of 50 robots, suffering nearly 50% packet loss, show that distributed estimation and control of image moments effectively achieves desired swarm formations.

  PublisherOA PDFDOI

• Human-Multirobot Collaborative Mobile Manipulation: the Omnid Mocobots

  arXiv (Cornell University) · 2022-06-28

  preprintOpen access

  The Omnid human-collaborative mobile manipulators are an experimental platform for testing control architectures for autonomous and human-collaborative multirobot mobile manipulation. An Omnid consists of a mecanum-wheel omnidirectional mobile base and a series-elastic Delta-type parallel manipulator, and it is a specific implementation of a broader class of mobile collaborative robots ("mocobots") suitable for safe human co-manipulation of delicate, flexible, and articulated payloads. Key features of mocobots include passive compliance, for the safety of the human and the payload, and high-fidelity end-effector force control independent of the potentially imprecise motions of the mobile base. We describe general considerations for the design of teams of mocobots; the design of the Omnids in light of these considerations; manipulator and mobile base controllers to achieve useful multirobot collaborative behaviors; and initial experiments in human-multirobot collaborative mobile manipulation of large, unwieldy payloads. For these experiments, the only communication among the humans and Omnids is mechanical, through the payload.

  PublisherOA PDFDOI

• Human-Multirobot Collaborative Mobile Manipulation: The Omnid Mocobots

  IEEE Robotics and Automation Letters · 2022 · 37 citations

  • Computer Science
  • Computer Science
  • Human–computer interaction

  The Omnid human-collaborative mobile manipulators are an experimental platform for testing control architectures for autonomous and human-collaborative multirobot mobile manipulation. An Omnid consists of a mecanum-wheel omnidirectional mobile base and a series-elastic Delta-type parallel manipulator, and it is a specific implementation of a broader class of mobile collaborative robots (“mocobots”) suitable for safe human co-manipulation of delicate, flexible, and articulated payloads. Key features of mocobots include passive compliance, for the safety of the human and the payload, and high-fidelity end-effector force control independent of the potentially imprecise motions of the mobile base. We describe general considerations for the design of teams of mocobots; the design of the Omnids in light of these considerations; manipulator and mobile base controllers to achieve multirobot collaborative behaviors; and experiments in human-multirobot collaborative mobile manipulation of large and articulated payloads, where the mocobot team renders the payloads weightless for effortless human co-manipulation. In these experiments, the only communication among the humans and Omnids is mechanical, through the payload.

  PublisherDOI

• Proximal Methods for Self-Healing and Exact Distributed Convex Optimization Extended abstract

  2022-09-27

  article1st authorCorresponding

  Distributed convex optimization algorithms employ a variety of methods for achieving exact convergence to the global optimal value (modulo numerical precision): some use time-varying dynamics, some use dynamics on each edge rather than on each node of the communication graph, some use double the communication between nodes per optimization step, and some use a specific initialization that enforces the dynamics to evolve on a particular subspace. Each of these methods has its drawbacks. Using time-varying dynamics might require a global clock, and it might cause transients due to disturbances to take longer and longer to die out as time progresses. Using edge-based rather than node-based dynamics might increase the memory and computation costs, and it typically requires the communication graph to be undirected. Using double the communication per optimization step might increase the communication cost, and it might also slow down the convergence to the optimal value. Using a specific initialization to enforce a particular invariant subspace might render the algorithm unable to recover from faults or disturbances that perturb its dynamics from this subspace, resulting in convergence to the incorrect value. In this latter case we say that the algorithm is not self-healing. In our previous work [1] we considered strongly convex objective functions having Lipschitz continuous gradients, and we introduced a new first-order method for achieving exact convergence to the global optimal value. Our new algorithm has none of the above drawbacks, and in particular it is a self-healing algorithm. But, it does possess a peculiar internal instability: each node has states that grows linearly in time even as its output converges to the optimal value. In the present work, we consider general non-smooth and extended-real-valued convex objective functions (which can thus incorporate hard convex constraints). We present proximal algorithms that again employ our new, internally unstable method for achieving exact convergence.

  PublisherDOI

• Randy A. Freeman [People in Control]

  IEEE Control Systems · 2022-11-18

  articleOpen access1st authorCorresponding

  <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Q. How did your education and early career lead to your initial and continuing interest in the control field?</b>

  PublisherOA PDFDOI

• On the Role of Well-Posedness in Homotopy Methods for the Stability Analysis of Nonlinear Feedback Systems

  Lecture notes in control and information sciences · 2021-09-11 · 2 citations

  book-chapter1st authorCorresponding
  PublisherDOI

Recent grants

• Decentralized nonlinear estimation and control of multi-agent systems

  NSF · $180k · 2006–2010

Frequent coauthors

• Kevin Lynch

  51 shared

• P.V. Kokotović

  23 shared

• Matthew L. Elwin

  Northwestern University

  15 shared

• Bryan Van Scoy

  11 shared

• Israel L. Donato Ridgley

  Northwestern University

  7 shared

• Brian Miller

  Merck & Co., Inc., Rahway, NJ, USA (United States)

  6 shared

• Jemin George

  6 shared

• P. Kokotovic

  University of California System

  6 shared

Labs

• Freeman LabPI