Observing the Effects of Overdesign in the Automatic Design of Control Software for Robot Swarms

2019

Swarm Intell

Self Cite

The reality gap-the discrepancy between reality and simulationis a critical issue in the off-line automatic design of control software for robot swarms, as well as for single robots. It is understood that the reality gap manifests itself as a drop in performance: when control software generated in simulation is ported to physical robots, the performance observed is often disappointing compared with the one obtained in simulation. In this paper, we investigate whether, to observe the effects of the reality gap, it is necessary to assume that the control software is designed in a context that is simpler than the one in which it is evaluated. In a first experiment, we show that a performance drop may be observed also in an artificial, simulation-only reality gap: control software is generated on the basis of a simulation model and assessed on a second one. We will call this second model a pseudo-reality. We selected the simulation model to be used as a pseudo-reality by trial and error, so as to qualitatively replicate previously published observations made in experiments with physical robots. The results show that a performance drop occurs even if we can exclude that pseudo-reality is more complex than the simulation model used for the design. In a second experiment, we eliminate the trial-and-error selection of the first experiment by evaluating control software across multiple pseudo-realities, which are sampled around the original simulation model used for the design. The results of the second experiment confirm those of the first one and show that they do not depend on the specific pseudo-reality we previously selected by trial and error. Moreover, they suggest that one could use multiple pseudo-realities to evaluate automatic design methods and, from this simulation-only evaluation, infer their robustness to the reality gap. The experiments were conceived by the two authors and performed by AL. The article was drafted by AL and revised by the two authors. The research was directed by MB.

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Simulation-only experiments to mimic the effects of the reality gap in the automatic design of robot swarms

Ligot

2019

Swarm Intell

Self Cite

“…EvoStick was formally defined in [18] to serve as a yardstick in the study of Vanilla, but had been previously analyzed in [19]. It was subsequently included in other empirical studies [17,16,5,33]. EvoCom targets the e-puck platform, as it is modeled by RM 2-see Table 1.…”

Section: Protocolmentioning

confidence: 99%

Automatic Design of Communication-Based Behaviors for Robot Swarms

Hasselmann

Robert

Lecture Notes in Computer Science

2018

Self Cite

We introduce Gianduja, an automatic design method that generates communication-based behaviors for robot swarms. Gianduja extends Chocolate, a previously published design method. It does so by providing the robots with the capability to communicate using one message. The semantics of the message is not a priori fixed. It is the automatic design process that implicitly defines it, on a per-mission basis, by prescribing the conditions under which the message is sent by a robot and how the receiving peers react to it. We empirically study Gianduja on three missions and we compare it with the aforementioned Chocolate and with EvoCom, a rather standard evolutionary robotics method that generates communication-based behaviors. We evaluate the behaviors produced by the three automatic design methods on a swarm of 20 e-puck robots. The results show that Gianduja uses communication meaningfully and effectively in all the three missions considered. The aggregate results indicate that, on the three missions considered, Gianduja performs significantly better than the two other methods under analysis.

“…The relative nature of performance drops might result in what we shall call a rank inversion: control software A outperforms control software B in simulation, but B outperforms A when assessed on the physical robots. Rank inversions can be observed when comparing instances of control software produced by different design methods [19], or by the same one at different steps along the optimization process [4]. Indeed, Birattari et al [4] observed a phenomenon that they called overdesign: past an optimal number of steps of the optimization process, the performance obtained in reality diverges from the one obtained in simulation.…”

Section: Introductionmentioning

confidence: 99%

On Mimicking the Effects of the Reality Gap with Simulation-Only Experiments

Ligot

Lecture Notes in Computer Science

2018

Self Cite

One issue in the automatic design of control software for robot swarms is the so-called reality gap-the difference between reality and the simulation models used in the automatic design process. It is commonly understood that the reality gap manifests itself as a drop in performance when control software developed in simulation is used to control physical robots. Yet, often disregarded is the relative nature of this performance drop: the reality gap does not affect equally all instances of control software. Indeed, one might observe a rank inversion: control software A might perform better than control software B in simulation, but perform worse on robots. The possibility of rank inversion undermines any performance comparison made in simulation. It would thus seem the only way to assess control software is in robot experiments, which are costly and time consuming. We argue it is unnecessary to assume reality is more complex than simulation models for the effects of the reality gap to occur. Indeed, we show that performance drop and rank inversion can occur if one automatically designs control software in simulation using a model and then assesses it in simulation on another model-what we call a pseudo-reality. Our results suggest that an appropriately conceived pseudo-reality could be used to test automatically-generated control software for performance drop and rank inversion, without performing robot experiments.