Warren Dixon, Professor of Mechanical and Aero Engineering, U. of Florida
| Jan/15 | Wed | 11:00AM-12:00PM | 4-149 |
Enrollment: Unlimited: No advance sign-up
Prereq: none
Analytical solutions to the infinite horizon optimal control
problem for continuous time nonlinear systems are generally not possible
because they involve solving a nonlinear partial differential equation.
Another challenge is that the optimal controller includes exact knowledge
of the system dynamics. Motivated by these issues, researchers have
recently used reinforcement learning methods that involve an actor and a
critic to yield a forward-in-time approximate optimal control design. This
presentation describes a forward-in-time dynamic programming approach that
exploits the use of concurrent learning tools where the adaptive update
laws are driven by current state information and recorded state information
to yield approximate optimal control solutions without the need for ad hoc
probing. Applications are presented for autonomous systems including robot
manipulators, underwater vehicles, and fin controlled cruise missiles.
Solutions are also developed for networks of systems where the problem is
cast as a differential game where a Nash equilibrium is sought.
Description of speaker: Warren Dixon is a Professor at the University of
Florida in the Mechanical and Aerospace Engineering Department and has published 3 books,
an edited collection, 9 chapters, and over 250 refereed journal and
conference papers and has received numerous awards for his work.
Sponsor(s): Electrical Engineering and Computer Science, Institute of Electrical and Electronic Engineers
Contact: Quanquan Liu, quanquan@mit.edu