Automatic Off-Line Design of Robot Swarms: A Manifesto

Abstract

Designing collective behaviors for robot swarms is a difficult endeavor due to their fully distributed, highly redundant, and ever-changing nature. To overcome the challenge, a few approaches have been proposed, which can be classified as manual, semi-automatic, or automatic design. This paper is intended to be the manifesto of the automatic off-line design for robot swarms. We define the off-line design problem and illustrate it via a possible practical realization, highlight the core research questions, raise a number of issues regarding the existing literature that is relevant to the automatic off-line design, and provide guidelines that we deem necessary for a healthy development of the domain and for ensuring its relevance to potential real-world applications.

Machine learning for active matter

“…Strategies found from simulation data usually work less well in the real world-this is the so-called reality gap 21,22,92 . Research on how to best avoid or overcome the reality gap is ongoing 93 .…”

“…Therefore, some caution must be taken when incorporating machine-learning models into the scientific process of understanding active matter. Guidelines for how to apply machine learning have been compiled, taking into account the specific issues arising in different fields [93][94][95][96] . Below we give guidelines suitable for initial application of machine learning in active-matter systems.…”

Phenotypic Plasticity Provides a Bioinspiration Framework for Minimal Field Swarm Robotics

“…Designing emergent (Matarić, 1993) and adaptive (Matarić, 1995) group behaviors is challenging, and so one can use evolutionary optimization in simulation before deployment (Dorigo et al, 2004;Trianni, 2008;Hecker and Moses, 2015;Birattari et al, 2019). In this way, adaptation of behavior can be seen in task specialization, for example, as an effective group-level strategy (Ferrante et al, 2015), though its effectiveness is tuned to the particular simulated environment.…”

Peer Review #1 of "Concurrent design of control software and configuration of hardware for robot swarms under economic constraints (v0.1)"

“…Automatic design is an alternative approach to designing a swarm. In automatic design, the design problem is formulated as an optimization problem that is then solved with an optimization algorithm (Birattari et al, 2019). A design problem of a collective mission is expressed as an objective function, a mathematical equation that measures the performance of the robot swarm.…”