Self-Replicating Robotics. In this research a novel approach of growing a neural network topology to control a modular robot morphology to attain full fledged self replication by the process of fusion and fission in a simulated world of sophisticated physics is proposed. A new model, Reinforced Central Pattern Generated Focussed Time Delay Neural Network is developed and deployed over stable and dock-able cylindrical modular morphology. Novel maneuvering capabilities directed towards self-replication are also synthesized.