VLSI Implementation of Neural Network Based Emergent Behavior Model for Robot Control

This paper reports the VLSI implementation of NN (N eural N etwork) based emergent behavior model for high-speed robot control. Augmented FSM (F inite-S tate M achine) is considered to implement the emergent behavior. We performed a system level simulation using our proposed model. Then, we transformed the model to RTL (R egister-T ransfer L evel) for circuit simulation. In this study, we considered multiple-inputs and multiple-outputs NN. Our implementation method improves speed of execution and accuracy and compare the result with conventional neural network. For activation function in NN, we implemented sigmoid function with second order approximation to educe complexity. We used walking gesture of Kondo KHR-3HV robot to verify the model.