*4.1. Missions*

We conceived three missions in which robot swarms operate in arenas surrounded by modular RGB blocks: STOP, AGGREGATION, and FORAGING. STOP and AGGREGATION are adaptations we make from the equivalent missions proposed by Hasselmann et al. to study Gianduja [14]. FORAGING is an abstraction of a foraging task, in a *best-of-n* fashion [63]—similar to the experiment of Valentini et al. [64]. The performance of the swarm is evaluated according to a mission-dependent objective function. We selected these missions because we conjecture that, to successfully perform them, the robots need to identify, process and/or transmit color-based information. In all missions, the time available to the robots is *T* = 120 s. The RGB blocks are arranged in walls that display colors on a per-mission basis—each RGB block is 0.25 m length and can display the colors red, green and blue {*R*, *G*, *B*}. In the context of these missions, when we reference to colored walls we imply that the RGB blocks arranged in the wall display the named color—for example, *"the green wall"* stands for a wall in which the RGB blocks composing it display the color green. In the following, we describe the scenario, the objective function, and the role of the colors for each mission. Figure 2 shows the arenas for the three missions.

**Figure 2.** Set-up of the simulated (top) and real arena (bottom) for the missions STOP (left), AGGREGATION (center), and FORAGING (right). The images show an example of the initial position of the robots.
