Task-based configuration synthesis for modular robot

A modular robot system is introduced in this paper. The requirements for configuration synthesis are analyzed and classified as hard ones, soft ones and hard-soft ones. The evaluation function is constructed based on the weighted sum of the evaluation results of the specified design requirements. A one-lever Genetic Algorithm (GA) basing on an improved hybrid coding method is presented to carry out the configuration synthesis. In addition an example is given to demonstrate the effectiveness of this method.