Swarms of flying robots are a promising alternative to ground-based robots for search in indoor environments with advantages such as increased speed and the ability to fly above obstacles. However, there are numerous problems that must be surmounted including limitations in available sensory and on-board processing capabilities, and low flight endurance. This paper introduces a novel strategy to coordinate a swarm of flying robots for indoor exploration that significantly increases energy efficiency. The presented algorithm is fully distributed and scalable. It relies solely on local sensing and low-bandwidth communication, and does not require absolute positioning, localisation, or explicit world-models. It assumes that flying robots can temporarily attach to the ceiling, or land on the ground for efficient surveillance over extended periods of time. To further reduce energy consumption, the swarm is incrementally deployed by launching one robot at a time. Extensive simulation experiments demonstrate that increasing the time between consecutive robot launches significantly lowers energy consumption by reducing total swarm flight time, while also decreasing collision probability. As a trade-off, however, the search time increases with increased inter-launch periods. These effects are stronger in more complex environments. The proposed localisation-free strategy provides an energy efficient search behaviour adaptable to different environments or timing constraints.

A major challenge in studying social behavior stems from the need to disentangle the behavior of each individual from the resulting collective. One way to overcome this problem is to constructa model of the behavior of an individual, and observe whether combining many such individuals leads to the predicted outcome.Thiscanbeachievedbyusingrobots.Inthisreviewwediscussthestrengthsandweaknessesof such an approach for studies of social behavior. We find that robots -whether studied in groups of researchers from a range of fields, as it poses interesting questions from mechanistic and evolutionary viewpoints. Many approaches have been used to study social behavior. These approaches can be classified over a scale of "situatedness", which we define as the extent to which individuals are embedded in an environment that can be sensed and modified by those individuals (Varela et al., 1991;Clark, 1996). The situatedness spectrum ranges from abstract mathematicalmodelsononeendtofieldworkinnaturalhabitatsattheotherend (Fig.1).

The artificial life approach to evolutionary robotics is used as a fundamental framework for the development of a modular neural control of autonomous mobile robots. The applied evolutionary technique is especially designed to grow different neural structures with complex dynamical properties. This is due to a modular neurodynamics approach to cognitive systems, stating that cognitive processes are the result of interacting dynamical neuro-modules. The evolutionary algorithm is described, and a few examples for the versatility of the procedures are given. Besides solutions for standard tasks like exploration, obstacle avoidance and tropism, also the sequential evolution of morphology and control of a biped is demonstrated. A further example describes the co-evolution of different neuro-controllers co-operating to keep a gravitationally driven art-robot in constant rotation

One of the key innovations during the evolution of life on earth has been the emergence of efficient communication systems, yet little is known about the causes and consequences of the great diversity within and between species. By conducting experimental evolution in 20 independently evolving populations of cooperatively foraging simulated robots, we found that historical contingency in the occurrence order of novel phenotypic traits resulted in the emergence of two distinct communication strategies. The more complex foraging strategy was less efficient than the simpler strategy. However, when the 20 populations were placed in competition with each other, the populations with the more complex strategy outperformed the populations with the less complex strategy. These results demonstrate a tradeoff between communication efficiency and robustness and suggest that stochastic events have important effects on signal evolution and the outcome of competition between distinct populations.

One advantage of the asynchronous and distributed character of embodied evolution is that it can be executed on real robots without external supervision. Further, evolutionary progress can be measured in real time instead of in generation based evaluation cycles. By combining embodied evolution with lifetime learning, we investigated a largely neglected aspect with respect to the common assumption that learning can guide evolution, the influence of maturation time during which an individual can develop its behavioral skills. Even though we found only minor differences between the evolution with and without learning, our results, derived from competitive evolution in predatorprey systems, demonstrate that the right timing of maturation is crucial for the progress of evolutionary success. Our findings imply that the time of maturation has to be considered more seriously as an important factor to build up empirical evidence for the hypothesis that learning facilitates evolution.