A mapping approach for controlling different maritime cranes and robots using ANN

In [1], a flexible and general control system architecture that allows for modelling, simulation and control of different models of maritime cranes and, more generally, robotic arms was previously presented by our research group. Each manipulator can be controlled by using the same universal input device regardless of differences in size, kinematic structure, degrees of freedom (DOFs), body morphology, constraints and affordances. The architecture presented establishes the base for the research of a flexible mapping procedure between a universal input device and the manipulators to be controlled, which is the topic of this paper. Based on the same architecture, as a validating case study, a new method for implementing such a mapping algorithm is introduced in this paper. This method is based on the use of Artificial Neural Networks. Using this approach, the system is able to automatically learn the inverse kinematic properties of different models. Learning is done iteratively based only on observation of input-output relationship, unlike most other control schemes. Related simulations are carried out to validate the efficiency of the proposed mapping method.