MetadataShow full item record
This Master?s thesis implements a multiple paired models architecture that isused to control a simulated robot. The architecture consists of several modules.Each module holds a paired forward/inverse model. The inverse model takes asinput the current and desired state of the system, and outputs motor commandsthat will achieve the desired state. The forward model takes as input the currentstate and the motor commands acting on the environment, and outputs thepredicted next state. The models are paired, due to the fact that the outputof the inverse model is fed into the forward model. A weighting mechanismbased on how well the forward model predicts determines how much a modulewill influence the total motor control. The architecture is a slight tweak of theHAMMER and MOSAIC architectures of Demiris and Wolpert, respectively.The robot is to imitate dance moves that it sees. Three experiments aredone; in the first two the robot imitates another robot, whereas in the thirdexperiment the robot imitates a movement pattern gathered from human data.The pattern was obtained using a Pro Reflex tracking system. After trainingthe multiple paired models architecture, the performance and self-organizationof the different modules are analyzed. Shortcomings with the architecture arepointed out along with directions for future work.The main results of this thesis is that the architecture does not self-organizeas intended; instead the architecture finds its own way to separate the inputspace into different modules. This is also most likely attributed to a problemwith the learning of the responsibility predictor of the modules. This problemmust be solved for the architecture to work as designed, and is a good startingpoint for future work.