Full-resolution Reconfiguration Planning for Heterogeneous Cube-shaped Modular Robots with only Sliding Motion Primitive

Hiroshi Kawano

Linear Version was presented @ ICRA2020 Virtual Conference. (available from IEEE Xplorer)

Quadratic version was presented @ ICRA2016. (available from IEEE Xplorer)

Abstract: This paper presents a full-resolution reconfiguration algorithm for a heterogeneous modular robot composed of sliding cube modules with a limited motion primitive. We assume that a cube module is not allowed to make convex motions around other modules but is allowed to slide across other modules’ surfaces. The availability of reconfiguration algorithm for such a module with limited mobility overcomes restrictions in the design of modular robots. To overcome the limited mobility of the assumed cube module, we introduce a method that compresses a robot with virtual walls and executes a reconfiguration and heterogeneous permutation of the robot structure in the compressed configuration. We prove the correctness and completeness of the proposed algorithm for three-dimensional connected structures. Simulation results show that the reconfiguration in free space is executed in quadratic operating time cost.

Background of the Research:

Previously studied reconfiguration methods for sliding cubic modular robot assume the availability of convex motion that provides the module high mobility. The cubic module that uses convex motion can go to any places on the surface of the robot structure. This guarantees the existence of movable module in the robot structure; here, the term “movable” is used when a modules can be removed from the robot structure without disconnecting the robot structure. But, if the cubic module is not allowed to use convex motion, the existence of the movable module is not guaranteed. And it causes a lot of difficulty in the reconfiguration of sliding cubic modular robots. For example, 1-D liner structure contains no movable module, and reconfiguration between configurations with different dimension is unavailable.

Figure 1. Limited accessibility of sliding only cubic modules on robot surface.

Figure 2. Unavailability of reconfiguration of sliding only cube between configurations with different dimension.

Summary of the proposed reconfiguration algorithm is as follows:

The proposed method consists of four stages:

1. Compression to the intermediate configuration.

2. Transformation to the next intermediate configuration that allows permutation process.

3. Permutation process.

4. Transformation to the intermediate configuration that is extended to the goal configuration in the next stage.

5. Extension to the goal configuration.

1. A robot is compressed to intermediate configuration with two perpendicular planes.

・Compression in two direction via virtual wall without disconnection.

　

2. The robot is transformed to the form with a rectangular ceiling in which the permutation process is carried out.

・Definition of intermediate configuration

・Module transfer between two perpendicular planes using seed modules

　　　

3. The permutation process is carried out.

・2x2 Rectangular space on the ceiling is used in each position exchange process.

4. As in 2., the robot is transformed to the form in which the robot has been prepared to be extended to the goal configuration.

5. The robot is extended to the goal configuration. (Just as the reverse motion of compression in (1) )

Full Demonstration:

Last Updated on 2018.06.14