Matthew Kelly
Controls Engineer at Boston Dynamics
Boston, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolMatthew Kelly is a controls-focused robotics engineer with 12 years of experience building real-time whole-body motion software for dynamic legged systems, currently driving controls for Boston Dynamics' Atlas humanoid. He holds a PhD in trajectory optimization and control from Cornell and has applied those techniques both in academic bipedal locomotion experiments and in production robot controllers. Prior roles at Rethink Robotics and MIT sharpened his path-planning, trajectory generation, and FPGA/embedded control skills, while his open-source OptimTraj library demonstrates hands-on expertise in implementing collocation methods and dynamics for trajectory optimization. Based in Boston, he blends deep theoretical grounding with practical implementation—often shipping robust, tested control code for demanding hardware.
12 years of coding experience
4 years of employment as a software developer
Mechanical Engineering, Mechanical Engineering at University of CanterburyBSME, Mechanical Engineering, BSME, Mechanical Engineering at Tufts UniversityDoctor of Philosophy (Ph.D.), Mechanical Engineering, Doctor of Philosophy (Ph.D.), Mechanical Engineering at Cornell University
English
Stackoverflow
431reputation
63kreached
5answers
4questions
Github Skills (11)
control-systems10
trajectory-optimization10
matlab10
numerical-methods10
algorithm9
algorithms9
numerical-integration8
casting6
eigen6
file-writing6
optimization6
Programming languages (7)
Github contributions (5)
MatthewPeterKelly/OptimTraj
Jun 2015 - Dec 2022
A trajectory optimization library for Matlab
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1 release, 7 reviews, 193 commits in 7 years 7 months
Contributions summary:Matthew's primary contribution involves implementing a trajectory optimization library for Matlab, as indicated by the repository description. The commits demonstrate the development of core functionalities within the library, including the derivation and implementation of dynamics, constraints, and objective functions for the cart-pole and five-link biped examples. They are focused on building a functional framework for trajectory optimization, integrating different methods such as trapezoid and Hermite-Simpson collocation, and testing with various cost functions.
matlaboptimizationrockettrajectorytrajectory-optimization
MatthewPeterKelly/dscTutorials
Aug 2014 - Dec 2015
Contributions:152 commits, 80 pushes, 2 branches in 1 year 4 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.