A new Torobo humanoid robot controlled by MTRNN with using force feedback control can interact with human dynamically. We are examining various emergent phenomena observed in the human-in-the-loop experiment.
Creative Dance MTRNN-PB
Humanoid Robot OP2 uses taught dances to create own new dance patterns.
First, tutors teach OP2 13 different dances. Those dances were used to train a MTRNN with parametric bias (PB). By exploring OP2’s PB space, the newly created dance patterns can be play back.
Two humanoid robots learn to predict each other’s behavior by inferring its intention. This project has been just started…
A humanoid robot learns to acquire a set of primitive behaviors and their combinations. The robot experiment utilizes MTRNN model which is characterized by its composition of fast and slow dynamics parts.
A humanoid robot spontaneously generates sequences of learned primitives. Chaos self-organized in the higher level of artificial brain (MTRNN model) generates pseudo stochastic sequences of moving an object among left, middle and right positions on a table.
A mobile robot with a hand learns to associate primitive sentences and corresponding behaviors with certain level of generalization. In the video, a robot, by recognizing a sentence “hit red”, generated the corresponding behavior. The robot was implemented with RNNPB model.