Study On Reconfiguration Plan Of Self-reconfigurable Modular Robot

With the extensive use of robot in the non-manufacturing industry, robot had played more and more important role in the working conditions of non-structure environment. The robot was required to have the ability of fitting the surroundings and changing tasks; nevertheless, owing to a close correlation between working ability and working performance of the robot, a dynamic changing self-configurable robot was needed. This kind of robot was called self-reconfigurable robot. In this paper, we research on the following aspects such as the module configuration, the configuration expression and the self-reconfiguration planning.Adaptability and self-reconfigurable capability of robot correlated closely with configuration of modular. Synthesizing advantages of M-Tran module designed by Murata et al. and HitMSR Module designed by Zhao Jie et al, a new self-reconfigurable robotic concept module was proposed, which had 2 DOFs and androgynous connect faces after researching deeply on module configurable characristic of home and abroad self-configurable robots.The number of DOFs and workspace of robot correlated closely with toplogy configuration of robot. Considering the characristic of module, A Deflexion State Assemble Incidence Matrix (DSAIM) was proposed to express configuration of robot. An algorithm of enumerating non-isomorphic configurations of the robot was proposed by using this matrix as well as the module symmetry in 3D space and combination principle. Configuration bases of 2 modules robot and 3 modules robot were established by using the enumerating algorithm.The self-reconfiguration plan of robot correlated closely with position and pose of robotic module in 3D space. In order to accomplish conveniently the self-configuration of the robot, a 3D-space Expression Matrix was defined which can express the relative position and state of each module of robot in space. And a kind of map algorithm was designed, which is from deflexion state assemble incidence matrix to 3D-space Expression Matrix. Then, by using this map algorithm, it can solve the judgement of the self-configuration of arbitrary topologic configuration and get the judgement algorithm. According to the decision algorithm of self-reconfiguration, we can get sets of the self-reconfigurable configuration by categorizing the sets of non-isomorphic configuration. We can express all the configurations and reconfiguration relations among them by using the Configuration Reconfiguration Incidence Map (CRIM), and then look for the best path of reconfigurable with the help of graph theory method. Then, combined with the reconfigurable processes which get before, we can realize the plan of hierarchical reconfigurable for the self-reconfigurable robot. Finally, the validity of the algorithm of configuration plan was verified by taking 3-module robot as an example.