Supervised Morphogenesis - Morphology Control of Ground-based Self-Assembling Robots by Aerial Robots

Supervised Morphogenesis - Morphology Control of Ground-based Self-Assembling Robots by Aerial Robots

Nithin Mathews, Alessandro Stranieri, Alexander Scheidler, Marco Dorigo

Abstract

In this paper, we study a heterogeneous robot team composed of self-assembling robots and aerial robots that cooperate with each other to carry out global tasks. We introduce supervised morphogenesis -- an approach in which aerial robots exploit their better view of the environment to detect tasks on the ground that require self-assembly, and perform on-board simulations to determine the morphology most adequate to carry out the task. In case existing morphologies on the ground do not match those determined in simulation, aerial robots use a series of enabling mechanisms to initiate and control (hence supervise) the formation of morphologies more adequate to carry out the task. Supervised morphogenesis solely employs LEDs and camera-based local communication between the two robot types. We validate the applicability of our approach in a real-world scenario, in which ground-based robots are given the task to cross an unknown, undulating terrain by forming ad-hoc morphologies under the supervision of an aerial robot.

Methodology

The eye-bot uses its downward-pointing camera from an elevated position to build an internal representation of the environment. In particular, two sequentially taken images are used to compute a height map of the environment in the field of view. The height map is used to calculate height profiles along each foot-bot's estimated trajectory to the light source. Subsequently, on-board simulations are performed to estimate whether each foot-bot is able to drive over the computed height profile of its estimated trajectory without toppling over. In case the simulation predicts a foot-bot to topple over, the eye-bot supervises the formation of target morphologies that offer the physical stability required to cross the hill. To initiate morphology formation, the eye-bot selects a favorably situated foot-bot. The eye-bot then establishes a dedicated one-to-one communication link to the selected foot-bot [1]. The dedicated communication link is then used to initiate the formation of a target morphology by activating the execution of a SWARMORPH-script [2]. SWARMORPH-script is a language that permits arbitrary morphology generation using self-assembling robots in a distributed manner. The foot-bots are pre-loaded with two different SWARMORPH-scripts that, when executed, can generate a chain morphology composed of two or three foot-bots each. Physical connections between a connection inviting foot-bot and a neighboring foot-bot are formed using the recruitment and guidance based mechanism presented in [3]. The robot controllers have been developed while following a distributed control paradigm.

Results

This is the video footage of an experiment we carried out to validate the real-world applicability of our approach. We use a hill with a maximum angle of inclination of 30°. Individual foot-bots are not able to cross the hill due to the too steep slope. A successfull task completion requires the assistance of the eye-bot to determine how many realizations of each target morphology (i.e., chain morphology of size two and three) are necessary. The eye-bot is kept stationary at all times. Therefore, the eye-bot uses a pre-calculated height map that was computed using images taken prior to running the experiment.

References

[1] N. Mathews, A. L. Christensen, E. Ferrante, R. O'Grady, and M. Dorigo. Establishing spatially targeted communication in a heterogeneous robot swarm. In Proceedings of 9th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2010), pages 939-946. IFAAMAS, 2010.
[2] A. L. Christensen, R. O'Grady, and M. Dorigo. SWARMORPH-script: a language for arbitrary morphology generation in self-assembling robots. Swarm Intelligence, 2(2-4):143-165, 2008.
[3] N. Mathews, A. L. Christensen, R. O'Grady, P. Rétornaz, M. Bonani, F. Mondada, and M. Dorigo. Enhanced Directional Self-Assembly Based on Active Recruitment and Guidance. In Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2011), pages 4762-4769. IEEE Computer Society Press, Los Alamitos, CA, 2011.