Emanuel (Emo) Vassilev Todorov (born 1971) is a neuroscientist and associate professor at the University of Washington, where he directs the Movement Control Laboratory. He is recognized for introducing optimal control as a formal explanatory framework for biological movement, a perspective that has become dominant in computational motor control. Todorov is also the principal developer of the MuJoCo physics engine, widely used in robotics and Machine learning research.
Todorov completed his PhD at the Massachusetts Institute of Technology under the supervision of Michael Jordan and Whitman Richards. He then held a postdoctoral fellowship at the Gatsby Computational Neuroscience Unit at University College London, working with Peter Dayan and Geoffrey Hinton. In 2004, he received a Sloan Fellowship in neuroscience.
Optimal Control and Motor Behavior
In 2002, Todorov proposed that stochastic optimal control principles provide a robust theoretical framework for explaining biological movement. This idea, initially met with skepticism, gained widespread acceptance over the following decade. In 2011, Karl Friston, a former critic, acknowledged that optimal control had become "the dominant paradigm for understanding motor behavior in formal or computational terms." The approach has been discussed in the popular scientific press alongside other connections between biology and optimization principles. An editorial by Kenji Doya in the Proceedings of the National Academy of Sciences described Todorov's work as "a refreshingly new approach in optimal control based on a novel insight as to the duality of optimal control and statistical inference."
Robotics and MuJoCo
Todorov's research extends to robotic hands, and his work has been featured in popular robotics publications. In January 2017, he was interviewed for the Robots Podcast, discussing his contributions to the field. The MuJoCo physics engine, developed under his leadership, has become a standard tool for simulating contact-rich dynamics in robotics and reinforcement learning, enabling advances in artificial intelligence research.
Academic Career and Funding
At the University of Washington, Todorov leads the Movement Control Laboratory, which investigates the neural and computational principles underlying motor control. His research has attracted substantial support, including 11 National Science Foundation grant awards totaling more than $7.5 million as principal investigator. This funding has supported projects spanning computational neuroscience, robotics, and deep learning applications.
Influence and Legacy
Todorov's integration of optimal control with biological movement has influenced both neuroscience and machine learning, particularly in areas such as model-based reinforcement learning and simulation-based robotics. His work bridges theoretical insights with practical tools, shaping how researchers model and control complex systems. As of the early 2020s, MuJoCo remains a widely used open-source platform, and his publications continue to be cited across disciplines.