Controlling Robots with Fractal Gene Regulatory Networks

Fractal proteins are a new evolvable method of mapping genotype to phenotype through a developmental process, where genes are expressed into proteins comprised of subsets of the Mandelbrot set. The resulting network of gene and protein interactions can be designed by evolution to produce specific patterns, which in turn can be used to solve problems. This chapter introduces the fractal development algorithm in detail and describes the use of fractal gene regulatory networks for learning a robot path through a series of obstacles. The results indicate the ability of this system to learn regularities in solutions and automatically create and use modules.

Perspectives for Strong Artificial Life

This chapter introduces the twin deadlocks of strong artificial life. Conceptualization of life is a deadlock both because of the existence of a continuum between the inert and the living, and because we only know one instance of life. Computationalism is a second deadlock since it remains a matter of faith. Nevertheless, artificial life realizations quickly progress and recent constructions embed an always-growing set of the intuitive properties of life. This growing gap between theory and realizations should sooner or later crystallize in some kind of “paradigm shift” and then give clues to break the twin deadlocks.

Generalized External Optimization

In this chapter a recently proposed meta-heuristic devised to be used in complex optimization problems is presented. Called Generalized Extremal Optimization (GEO), it was inspired by a simple co-evolutionary model, developed to show the emergence of self-organized criticality in ecosystems. The algorithm is of easy implementation, does not make use of derivatives and can be applied to unconstrained or constrained problems, non-convex or even disjoint design spaces, with any combination of continuous, discrete or integer variables. It is a global search meta-heuristic, like the Genetic Algorithm (GA) and the Simulated Annealing (SA), but with the advantage of having only one free parameter to adjust. The GEO has been shown to be competitive to the GA and the SA in tackling complex design spaces and a useful tool in real design problems. Here the algorithm is described, including a step-by-step implementation to a simple numerical example, its main characteristics highlighted, and its efficacy as a design tool illustrated with an application to satellite thermal design.

Integrating Evolutionary Computation Components in Ant Colony Optimization

This chapter introduces two different ways to integrate Evolutionary Computation Components in Ant Colony Optimization (ACO) Meta-heuristic. First of all, the ACO meta-heuristic is introduced and compared to Evolutionary Computation to notice their similarities and differences. Then two new models of ACO algorithms that include some Evolutionary Computation concepts (Best-Worst Ant System and exchange of memoristic information in parallel ACO algorithms) are presented with some empirical results that show improvements in the quality of the solutions when compared with more basic and classical approaches.

Genetic Programming Using a Turing-Complete Representation

In this chapter, a new representation scheme for Genetic Programming (GP) is proposed. We need a Turing-complete representation for a general method of generating programs automatically; that is, the representation must be able to express any algorithms. Our representation is a recurrent network consisting of trees (RTN), which is proved to be Turing-complete. In addition, it is applied to the tasks of generating language classifiers and a bit reverser. As a result, RTN is shown to be usable in evolutionary computing.

The Influence of Pheromone and Adaptive Vision in the Standard Ant Clustering Algorithm

Algorithms inspired by the collective behavior of social organisms, from insect colonies to human societies, promoted the emergence of a new field of research called swarm intelligence. The applications of swarm intelligence range from routing in telecommunication networks to robotics. This chapter discusses some of the ideas behind swarm intelligence, focusing on a clustering algorithm motivated by the social behavior of some ant species. The standard ant-clustering algorithm is presented; a brief review from the literature concerning the applications and variations of the basic model is provided; two novel modifications of the original algorithm are proposed and discussed; and a sensitivity analysis of the standard and modified algorithm in relation to some user-defined parameters is performed. A variation of a simple benchmark problem in the field is used to perform the sensitivity analysis of the algorithm and to assess the proposed modifications of the standard algorithm.

From Biologically Inspired Computing to Natural Computing

Biologically inspired computing is just one of the branches of natural computing, which also encompasses artificial life, fractal geometry and computing with natural means (molecular, membrane and quantum computing). This chapter provides a brief and general overview of natural computing, focusing on bio-inspired algorithms. Some relevant literature is cited for guidance purposes and the main objective and scope of the book is described.

Continuous Attractor Neural Networks

In this chapter a brief review is given of computational systems that are motivated by information processing in the brain, an area that is often called neurocomputing or artificial neural networks. While this is now a well studied and documented area, specific emphasis is given to a subclass of such models, called continuous attractor neural networks, which are beginning to emerge in a wide context of biologically inspired computing. The frequent appearance of such models in biologically motivated studies of brain functions gives some indication that this model might capture important information processing mechanisms used in the brain, either directly or indirectly. Most of this chapter is dedicated to an introduction to this basic model and some extensions that might be important for their application, either as a model of brain processing, or in technical applications. Direct technical applications are only emerging slowly, but some examples of promising directions are highlighted in this chapter.

Biologically Inspired Collective Robotics

In this chapter, we review our recent research in the area of collective robotics, and the problem of controlling multiple robots in the completion of common tasks. Our approach is characterized with a strong inclination for biological inspiration in which examples in nature — social insects in particular — are used as a way of designing strategies for controlling robots. This approach has been successfully applied to the study of three representative tasks, namely, collective box-pushing, collective construction, and collective sorting. Collective box-pushing deals with the purposeful motion of an object too large to be moved by a single robot and we rely on the group prey transport phenomenon found in ants to derive the necessary behaviors for accomplishing this task. Collective construction is concerned with the building of a geometric structure with the combined efforts of many individuals in parallel, without centralized control and we study a species of ant known to possess this capability, to model and control the process of creating a circular nest with multiple robots. Finally, in collective sorting the broad behavior in ants serves as the motivation behind designing robotic behaviors that depend on only local sensing in clustering objects of different types into separate piles. The success of our proposed approach is supported by both simulation and physical experiments using robots.

Particle Swarms

Particle swarm optimization is a computer paradigm that is based on human social influence and cognition. Candidate problem solutions are randomly initialized, and improvements are found through interactions among them. Social-psychological aspects of the algorithm are described, followed by implementation details. The particle swarm operates in three kinds of spaces, namely a topological space comprising the “social network” structure of the population, a parameter space of problem variables, and a one-dimensional evaluative space. Variations in the algorithm are described, and finally it is compared to evolutionary computation models.