This has been my very first research topic since the time of my PhD studies at IRIDIA. My thoughts about evolutionary robotics (ER) are well described in a recent paper published on Frontiers in Robotics and AI.
My research in ER is mainly targeted to the engineering of complex robotics systems, particularly within the swarm robotics domain. In the attempt to define a general methodology that goes beyond task-specific solutions, I have proposed to investigate the effects of the several design choices that must be made when adopting the ER approach. These design choices pertain the specific aspects of an evolutionary design process specifically targeted to robotics and adaptive behaviours, and can be classified in four macro classes:
- The robot morphology and sensorimotor configuration constitutes the interface between the external world and the control system. The correct engineering of the robot configuration can lead to improved performance (Trianni and Nolfi, 2011; Fehérvári et al., 2013).
- The genotype-to-phenotype mapping represents the link between the evolutionary algorithm and the robotic system, and clearly influences the efficiency of the evolutionary optimisation.
- The fitness function and the evolutionary algorithm determine the way in which potential solutions are retained or discarded, and how the search space is explored. In ER, much attention has been dedicated to the definition of the fitness function, while algorithms have been mostly mediated from research in evolutionary computation. In this respect, I propose the use of Multi=Objective Optimisation to (i) support the definition of the fitness by using multiple simple objectives, and (ii) enhance the ability to search the space of all potential solutions (Trianni and and López-Ibáñez, 2015)
- The ecological context represents the possible task variations that need to be tackled. This normally corresponds to varying starting positions of the robots, as well as varying parameters of the task environment. The way in which the ecological context is explored during the evolutionary optimization may produce selective pressures that have a bearing on the flexibility and robustness of the generated solutions.
Besides engineering, ER can be a powerful tool to model biological systems. In this context, ER offers the unique possibility to run evolutionary experiments in silico, where robotic organisms undergo a Darwinian selection process that shapes their morphology and behaviour within the given ecological context. In principle, ER allows us to identify the causal relationship between selective pressures and adaptive traits, thanks to the possibility of having complete control over the evolutionary process. Additionally, ER can be exploited to automatically generate animats with relevant behavioural and cognitive abilities. In this case, artificial evolution serves just as the optimisation process, and could in principle be replaced by any other method of synthesizing the animat. In other words, the modeling effort focuses on the animat itself, and not on the process of obtaining it.