The case for engineering the evolution of robot controllers

In this paper, we discuss how the combination of evolution-ary techniques and engineering-oriented approaches is an ef-fective methodology for leveraging the potential of evolu-tionary robotics (ER) in the synthesis of behavioural control. We argue that such combination can eliminate the issues that have prevented ER from scaling to complex real-world tasks, namely: (i) the bootstrap problem, (ii) deception, (iii) the re-ality gap effect, and (iv) the prohibitive amount of time nec-essary to evolve controllers directly in real robotic hardware. We present recent studies carried out in our research group involving real-robot and simulation-based experiments. We provide examples of how the synergistic effects of evolution and engineering overcome each other’s limitations and signif-icantly extend their respective capabilities, thereby opening a new path in the design of robot controllers