Having discovered a penchant for open-ended research during my time at the Michigan Aerospace Corporation, I returned to graduate school in 2020 as a PhD Student in Aerospace Engineering at the University of Michigan, where I now work in Professor Dimitra Panagou's Distributed Aerospace Systems and Control Lab. Focusing mainly on safety-critical systems, my interests include multi-agent networks and control and estimation under uncertainty, including both adaptive and learning-based techniques. A summary of my work may be found here. In the coming years, I aim to develop a propensity toward teaching and mentorship while I build my research portfolio in hopes of pursuing a career in academia.
When I graduated from Tufts University in 2016 I dove headfirst into a PhD program at Michigan, and I didn’t know what I didn’t know. I managed to tread water long enough to obtain my Master’s degree, but I left academia to pursue professional running and industry experience so that I could re-evaluate my goals. Once I regained my footing, I decided to make helping other graduate students who might be struggling a priority. Thus the idea for Dissertakes on a PhD was born. Over the coming months, I will be interviewing current, past, and future graduate students and documenting my findings in order to help paint a clearer picture of what life is like as a PhD student before, during, and after those elusive three letters are earned. I hope you will come to appreciate the various Takes and all that they offer for graduate students in all walks of life.
A firm believer in the interplay between a sound body and sound mind, I spend my breaks from research running hard intervals on the track or depriving my lungs of oxygen on some of Ann Arbor’s illustrious hills. After an unsuccessful attempt to qualify for the 2016 USA Track & Field Olympic Trials, I was able to sneak into the 2021 edition of the Trials at 800m with a PB of 1:47.43. Though no longer training at this level, the interested reader may visit my log and follow along for workouts, long runs, and the ambling musings of a middle-distance runner with more to prove.
Mitchell K. Black Jr.
Ph.D. Candidate at the University of Michigan
Driven by a desire to understand the human experience, I have made inroads in the field of safe control under uncertainty in both academia and industry at the confluence of system identification, adaptive control, and risk-aware control. I am eager to continue working on control, estimation, and learning for autonomous vehicle applications with a team of research-minded professionals.
SKILLS
Python, MATLAB/Simulink, C++, ROS, git, Carla, control barrier functions, learning-based control, linear and nonlinear system identification, nonlinear control, stochastic control, adaptive control, risk-aware control, neural networks for safe control, Koopman operator theory, convex optimization, finite- and fixed-time estimation and control, real-time robot control
HIGHLIGHTS
After developing a novel form of safe, predictive control for autonomous vehicle applications in my PhD research, I accepted a summer internship with Toyota aimed at investigating its application to risk-aware control. This culminated in a patent application, a paper submission to ICRA 2023 (under review), and an extension of the summer position through March 2023.
In the last 8 months I have been the first author on 4 conference paper submissions and 1 journal submission (all of which are under review) and served as Graduate Student Instructor for a senior undergraduate controls course during the Fall semester.
EDUCATION
Ph.D. in Aerospace Engineering | University of Michigan | Ann Arbor, MI 2020 – 2023
Dissertation: “Toward Safe Control under Uncertainty: Adaptation and Prediction for Control Barrier Functions”
GPA: 3.77 / 4.00 – Multi-agent systems, Nonlinear Systems and Control
M.S.E. in Aerospace Engineering | University of Michigan | Ann Arbor, MI 2016 – 2017
GPA: 3.75 / 4.00 – Linear Systems Theory, Navigation and Guidance, Trajectory Optimization
B.S. in Mechanical Engineering and Astrophysics | Tufts University | Medford, MA 2012 – 2016
GPA: 3.73 / 4.00, Magna cum laude
Awarded 2016 NCAA Walter Byers Scholarship (given to 1 male and 1 female nationwide) and awarded $48k for postgraduate study
Captain of Varsity Track & Field Team, 4x NCAA National Champion 800m Run, 10x NCAA Track & Field All-American
Awarded Tufts Ciaffone & Pote Scholarship for Excellence in Engineering
One of Twelve Recipients of 2016 Tufts University Senior Award
EXPERIENCE
CO-OP | Toyota Research Institute of North America | Ann Arbor, MI May 2022 – Present
Developing novel techniques for risk-aware control of safety-critical systems using control barrier functions and neural-network based methods in support of the cyber-physical systems team’s research objectives.
Submitted invention disclosure and patent application for novel controller
Submitted work for publication at the 2023 International Conference on Robotics and Automation (ICRA)
GRADUATE STUDENT INSTRUCTOR | University of Michigan | Ann Arbor, MI (Aug – Dec) 2021, 2022
Assisted students with “Control for Aerospace Vehicles” (senior-level undergraduate course) by preparing additional material, explaining theoretical concepts, and solving example problems during 20 hour per week Fall semester appointment.
Prepared and presented lecture on safe control design with control barrier functions (12/08/2022)
Presented lecture on the Bode stability criterion, including gain margin and phase margin (11/10/2022)
RESEARCH SCIENTIST | Michigan Aerospace Corporation | Ann Arbor, MI 2017 – 2020
Created and operated web application for collection of mission-critical atmospheric LiDAR data in support of rocket launches.
Built web application with Python/VueJS for control of LiDAR data collection devices measuring differential clear air turbulence
Increased collected LiDAR data by >100% by investigating and implementing search algorithms for laser clearinghouse compliant data collection scheduler
PROPULSION ACADEMY RESEARCH ASSOCIATE | NASA Marshall Space Flight Center | Huntsville, AL (Jun – Aug) 2016
Studied efficacy of parallel dynamic force feedback (DFF) / proportional-integral-derivative (PID) controller for resonance damping during rocket engine ignition.
Conducted experiments to model electromechanical actuator and pendulum system via Bode analysis
Optimized DFF / PID control gains via simulation in MATLAB and Simulink for demonstrating feasibility of controller for thrust vector control application
CERTIFICATIONS
DIVIDE AND CONQUER, SORT AND SEARCH, AND RAND. ALGS. | Stanford University via Coursera | Online Sep 2019
MACHINE LEARNING | Stanford University via Coursera | Online Apr 2019
RESEARCH
Fields of Interest
Safe autonomy, safety-critical systems, motion planning and control, decentralized control for multi-agent systems, risk-aware/risk-bounded control, adaptive control, learning-based control, predictive control, data-driven control, Koopman operator theory, stochastic systems, nonlinear systems
Active Areas of Research
Risk-aware control for stochastic safety-critical systems, online parameter adaptation for multiple control barrier functions, predictive control with control barrier functions (without model predictive control), nonlinear system identification and control for safety-critical systems, probabilistic set reachability with predictive control, recurrent neural networks for system identification
Conference Publications
Mitchell Black, Kunal Garg, and Dimitra Panagou, “A Quadratic Program based Control Synthesis under Spatiotemporal Constraints and Non-vanishing Disturbances”. 2020 IEEE Conference on Decision and Control.
Mitchell Black, Ehsan Arabi, and Dimitra Panagou, “A Fixed-Time Stable Adaptation Law for Safety-Critical Control under Parametric Uncertainty”. 2021 European Control Conference.
Mitchell Black and Dimitra Panagou. “Adaptation for Validation of a Consolidated Control Barrier Function based Control Synthesis.” 2023 IEEE International Conference on Robotics and Automation. (Under Review).
Mitchell Black, Georgios Fainekos, Bardh Hoxha, Danil Prokhorov, and Dimitra Panagou. “Safety under Uncertainty: Tight Bounds with Risk-Bounded Control Barrier Functions.” 2023 IEEE International Conference on Robotics and Automation. (Under Review).
Mitchell Black, Mrdjan Jankovic, Abhishek Sharma, Dimitra Panagou. “Future-Focused Control Barrier Functions for Autonomous Vehicle Control.” 2023 American Control Conference. (Under Review).
Mitchell Black and Dimitra Panagou. “Safe Control Design for Unknown Nonlinear Systems with Koopman-based Fixed-Time Identification.” 2023 IFAC World Congress. (Under Review).
Journal Articles
Mitchell Black, Ehsan Arabi, and Dimitra Panagou, “Fixed-Time Parameter Adaptation for Safe Control Synthesis.” Automatica. (Under Review).
Mitchell Black and Dimitra Panagou. “Online Adaptation for Provably Safe Control under Input Constraints and Multiple Control Barrier Functions.” (In Preparation).
ATHLETICS
STUDENT ATHLETE | Tufts University Varsity Track and Field/Cross Country | Medford, MA 2012 — 2016
Elected Team Captain unanimously by 60+ team members as a Senior
Won 4 NCAA DIII National Championships at 800m and earned 10 NCAA All-American Honors
Named 2016 Capital One Sports Information Directors of America Academic All-American of the Year
ATHLETE | Very Nice Track Club / Craftsbury Green Racing Project | Ann Arbor, MI 2018 — 2022
Qualified for the 2021 US Track & Field Olympic Trials at 800m with a personal best of 1:47.43 (placed 23rd in US)
Qualified for the 2018 US Indoor Track & Field Championships at 800m (placed 12th in US)