Oral
Interactive Control of Diverse Complex Characters with Neural Networks
Igor Mordatch · Kendall Lowrey · Galen Andrew · Zoran Popovic · Emanuel Todorov

Tue Dec 8th 04:30 -- 05:30 PM @ Room 210 A

We present a method for training recurrent neural networks to act as near-optimal feedback controllers. It is able to generate stable and realistic behaviors for a range of dynamical systems and tasks -- swimming, flying, biped and quadruped walking with different body morphologies. It does not require motion capture or task-specific features or state machines. The controller is a neural network, having a large number of feed-forward units that learn elaborate state-action mappings, and a small number of recurrent units that implement memory states beyond the physical system state. The action generated by the network is defined as velocity. Thus the network is not learning a control policy, but rather the dynamics under an implicit policy. Essential features of the method include interleaving supervised learning with trajectory optimization, injecting noise during training, training for unexpected changes in the task specification, and using the trajectory optimizer to obtain optimal feedback gains in addition to optimal actions.

Author Information

Igor Mordatch (University of Washington)
Kendall Lowrey (University of Washington)
Galen Andrew (University of Washington, Seattle)
Zoran Popovic (University of Washington)
Emanuel Todorov (University of Washington)

More from the Same Authors