Evolutionary robots with on-line self-organization and behavioral fitness

Abstract

We address two issues in Evolutionary Robotics, namely the genetic encoding and the performance criterion, also known as the fitness function. For the first aspect, we suggest to encode mechanisms for parameter self-organization, instead of the parameters themselves as in conventional approaches. We argue that the suggested encoding generates systems that can solve more complex tasks and are more robust to unpredictable sources of change. We support our arguments with a set of experiments on evolutionary neural controllers for physical robots and compare them to conventional encoding. In addition, we show that when also the genetic encoding is left free to evolve, artificial evolution will select to exploit mechanisms of self-organization. For the second aspect, we shall discuss the role of the performance criterion, als known as fitness function, and suggest Fitness Space as a framework to conceive fitness functions in Evolutionary Robotics. Fitness Space can be used as a guide to design fitness functions as well as to compare different experiments in Evolutionary Robotics.

Real-time Hebbian Learning from Autoencoder Features for Control Tasks

“…Oja 1982;Bienenstock et al 1982). Researchers also began to evolve such ANNs in the hope of achieving more brain-like functionalities by producing networks that change over their lifetime (Floreano and Urzelai, 2000;Niv et al, 2002;Risi and Stanley, 2010;Risi et al, 2011;Stanley et al, 2003).…”

“…As a medium for adaptation and learning, neural plasticity has long captivated artificial life and related fields (Baxter, 1992;Floreano and Urzelai, 2000;Niv et al, 2002;Soltoggio et al, 2008Soltoggio et al, , 2007Soltoggio and Jones, 2009;Soltoggio and Stanley, 2012;Risi and Stanley, 2010;Risi et al, 2011;Risi and Stanley, 2012;Stanley et al, 2003;Coleman and Blair, 2012). Much of this body of research focuses on Hebbian-inspired rules that change the weights of connections in proportion to the correlation of source and target neuron activations (Hebb, 1949).…”

“…Much of this body of research focuses on Hebbian-inspired rules that change the weights of connections in proportion to the correlation of source and target neuron activations (Hebb, 1949). The simplicity of such Hebbian-inspired rules makes them easy and straightforward to integrate into larger systems and experiments, such as investigations into the evolution of plastic neural networks (Floreano and Urzelai, 2000;Soltoggio et al, 2008;Risi et al, 2011). Thus they have enabled inquiry into such diverse problems as task switching (Floreano and Urzelai, 2000), neuromodulation (Soltoggio et al, 2008), the evolution of memory (Risi et al, 2011), and reward-mediated learning (Soltoggio and Stanley, 2012).…”

Evolution of Embodied Intelligence

“…Of course, evolution is also a form of adaptation, but modifications occur only over several generations, and that may require too long time for a robotic system (for a comparative discussion of lifelong learning and evolution, see [27]). In order to compensate for the problems of both approaches, we decided to genetically encode and evolve the mechanisms of neural adaptation [11]. The idea was to exploit evolution to find good combinations of learning structures, rather than static controllers, and to evolve learning structures that without the constraints of off-the-shelf learning algorithms.…”

Evolution of Heterogeneous Cellular Automata in Fluctuating Environments

“…In parallel to biological research, a number of studies in artificial life, especially evolutionary robotics (Floreano and Urzelai, 2000), have also investigated environmental variations, some of them explicitly defining the environment as a driving evolutionary force (Bredeche and Montanier, 2012). Others, such as Lipson et al (2002), showed a correlation between the modularity and the rate of change of external resources, while Yu (2007) observed that populations exploit neutrality to cope with environmental fluctuations and can evolve a type of evolvability under two alternating objective functions.…”