Multi-objective configuration optimization of assembly-level reconfigurable modular robots

Consisting of a collection of modules with the same connection interfaces, a reconfigurable modular robot can be assembled into different configurations to meet different task requirements. The configuration optimization of modular robot is formulated as an optimal design process that is selecting sets of modules and assembling them together to get optimal configurations. To solve the problem of configuration design, this paper presents a multi-objective optimization method for assembly-level modular robots which can bring us more convince to present the modular robots. An automatic generation method of the configuration from an assembly model is presented. With the assembly as the design variable, a multi-objective optimization model of configuration optimization is founded to search for the optimal solutions. Based on our modular robot system, an example is provided to verify the effectiveness of this approach.