Best Systems Paper, RSS 2016.
In this paper, we present an end-to-end system for addressing tasks with modular robots, and demonstrate that it is capable of accomplishing challenging multi-part tasks in hardware experiments.
The advantage of modular robot systems lies in their flexibility, but this advantage can only be realized if there exists some reliable, effective way of generating configurations (shapes) and behaviors (controlling programs) appropriate for a given task. Our goal in this work was to provide a system that enables users to solve complex tasks using modular robot hardware.
Our approach is library-driven: rather than attempting to generate new designs from scratch, users specify task requirements and a design tool retrieves designs satisfying the requirements from a library of existing useful designs. Our system integrates low-level design generation, library management, high level mission planning, and modular robot hardware.
At the low-level, the tedious, error-prone process of designing configurations and behaviors for a modular robot system is simplified by our growing modular robot design library. At the high-level, controller synthesis with formal language provides guarantees on the correctness of synthesized controllers. The seamless connection between the high-level and low-level aspects results in a useful end-to-end system for controlling modular robot systems to accomplish user specified tasks.
Our simulation and design tool for SMORES-EP can be downloaded for Windows, Mac, and Linux at www.vsparc.org
G. Jing, T. Tosun, M. Yim, and H. Kress-Gazit, “An end-to-end system for accomplishing tasks with modular robots,” in Robotics: Science and Systems, 2016. doi: 10.15607/RSS. 2016.XII.025. (PDF)