Toward Evolutionary Design of Autonomous Systems

Brain and Cognition, 1997

Recently there have been a number of proposals for the use of artificial evolution as a radically new approach to the development of control systems for autonomous robots. This paper explains the artificial evolution approach, using work at Sussex to illustrate it. The paper revolves around a case study on the concurrent evolution of control networks and visual sensor morphologies

The artificial evolution of intelligence is discussed with respect to current methods. An argument for withdrawal of the traditional `fitness function' in genetic algorithms is given on the grounds that this would better enable the emergence of intelligence, necessary because we cannot specify what intelligence is. A modular developmental system is constructed to aid the evolution of neural structures and a simple virtual world with many of the properties believed beneficial is set up to test these ideas. Resulting emergent properties are given, along with a brief discussion. Keywords: Artificial Intelligence, Emergence, Genetic Algorithms, Artificial Life, Neural Networks, Development, Modularity, Fractals, Lindenmayer Systems, Recurrence. ii Acknowledgments Thanks to my supervisor Inman Harvey for his encouragement in the area and comments on this project. iii CONTENTS 1 INTRODUCTION..................................................................................................

Proceedings. 1998 IEEE International Conference on Robotics and Automation (Cat. No.98CH36146), 1998

This paper explores the use of neural networks to control robots in tasks requiring sequential and leaming behavior. We propose a Family Competition Evolutionary Algorithm (FCEA) to evolve networks that can integrate these different types of behavior in a smooth and continuous manner. The approach integrates selfadaptive Gaussian mutation, self-adaptive Cauchy mutation, decreasing-based Gaussian mutation, and family competition. In order to illustrate the power of the approach, we apply this approach to two different task domains: the "artificial ant" problem and a sequential behavior probleman agent learns to play football. From the experimental results, we find our approach performs much better than other evolutionary algorithms in these two tasks. The main contribution of the paper, based on the results from our experiments, is that our approach can evolve neural networks to provide a means of integrating sequencing and learning within a single control system.

2008

Abstract- Natural deduction is essentially a sequential decision task, similar to many game-playing tasks. Such a task is well suited to benefit from the techniques of neuro-evolution. Symbiotic, Adaptive Neuro-Evolution (SANE)(Moriarty and Miikkulainen 1996) has proven successful at evolving networks for such tasks. This paper will show that SANE can be used to evolve a natural deduction system on a neural network. Particularly, it will show that (1) incremental evolution through progressively more challenging problems results in more effective networks than does direct evolution, and (2) an effective network can be evolved faster if the network is allowed to “brainstorm ” or suggest any move regardless of its applicability, even though the highest-ranked valid move is always applied. This way evolution results in neural networks with human-like reasoning behavior. 1

The aim of this paper is to present a biologically inspired Neuro Evolutive Algorithm (NEA) able to generate modular, hierarchical and recurrent neural structures as those often found in the nervous system of live beings, and that enable them to solve intricate survival problems. In our approach we consider that a nervous system design and organization is a constructive process carried out by genetic information encoded in DNA. Our NEA evolves Artificial Neural Networks (ANNs) using a Lindenmayer System with memory that implements the principles of organization, modularity, repetition (multiple use of the same sub-structure), hierarchy (recursive composition of sub-structures), as a metaphor for development of neurons and its connections. In our method, this basic neural codification is integrated to a Genetic Algorithm (GA) that implements the constructive approach found in the evolutionary process, making it closest to the biological ones. Our method was initially tested on a simp...

This paper outlines a preliminary step towards the long-term goal of intelligent artificial life. Evolutionary emergence via natural selection is proposed as the way forward, in combination with other biologically-inspired principles including the developmental modularity of neural networks. In order to develop and test the ideas, an artificial world containing autonomous organisms has been created. Its underlying theory and construction are described. Resulting emergent strategies are reported both from an observer's perspective and in terms of their neural mechanisms. The results prove that the proposed approach is viable and show it to be an exciting area for further research. 1 Introduction The interest of this work is the generation of intelligence worthy of comparison with that found in natural life, even if only at a rudimentary level. This requires the creation of systems that adapt to behave in `intelligent' ways within an environment, without being given any inform...

Evolving Artificial Neural Networks (ANN) is a new method that, except of the training, was applied to the structure optimization problem. This method combines ideas from both the evolution and adaptive signal processing techniques. An ANN is considered as a layered array of non-linear systems (the neuron models), each producing on its output a local error. Each of these errors is minimized using the Extended Kalman Filter (EKF). Then the evolutionary algorithm is used to search for the best array that minimizes the global error on the network's output. The initial population is randomly created consisting of ANN with different structure (in the hidden layer). The proposed algorithm has been tested with two different real world applications giving very promising results.

Cognitive Technologies, 2007

This chapter introduces the idea of "Evolvable Hardware," which applies evolutionary algorithms to the generation of programmable hardware as a means of achieving Artificial Intelligence. Cellular Automata-based Neural Networks are evolved in different modules, which form the components of artificial brains. Results from past models and plans for future work are presented.

Understanding Complex Systems, 2009

Organic computing calls for efficient adaptive systems in which flexibility is not traded in against stability and robustness. Such systems have to be specialized in the sense that they are biased towards solving instances from certain problem classes, namely those problems they may face in their environment. Nervous systems are perfect examples. Their specialization stems from evolution and development. In organic computing, simulated evolutionary structure optimization can create artificial neural networks for particular environments. In this chapter, trends and recent results in combining evolutionary and neural computation are reviewed. The emphasis is put on the influence of evolution and development on the structure of neural systems. It is demonstrated how neural structures can be evolved that efficiently learn solutions for problems from a particular problem class. Simple examples of systems that "learn to learn" as well as technical solutions for the design of turbomachinery components are presented.