Disentangling automatic and semi-automatic approaches to the optimization-based design of control software for robot swarms

Optimization-based design is an effective and promising approach to realizing collective behaviours for robot swarms. Unfortunately, the domain literature often remains vague about the exact role played by the human designer, if any. It is our contention that two cases should be disentangled: semi-automatic design, in which a human designer operates and steers an optimization process (for example, by fine-tuning the parameters of the optimization algorithm); and (fully) automatic design, in which the optimization process does not involve, need or allow any human intervention. In this Perspective, we briefly review the relevant literature; illustrate the hypotheses, characteristics and core challenges of semi-automatic and automatic design; and sketch out the context in which they could be ideally applied.