Vision Controlled Humanoid Robot Tool-Kit

This paper introduces a novel parallelised vision based intelligent controller for a Humanoid Robot system. This intelligent controller is simulated dynamically and its performance evaluated for a standard benchmark problem. The parallel nature of the simulation architecture which can separate the image processing and control algorithms allows the simulation to progress in real-time or faster than real-time. This allows automated control algorithms using neural network or evolutionary algorithms to be efficiently and effectively developed.