José Santos

Multimodule Artificial Neural Network Architectures for Autonomous Robot Control Through Behavior Modulation

In this paper we consider one of the big challenges when constructing modular behavior architectures for the control of real systems, that is, how to decide which module or combination of modules takes control of the actuators in order to implement the behavior the robot must perform when confronted with a perceptual situation. The problem is addressed from the perspective of combinations of ANNs, each implementing a behavior, that interact through the modulation of their outputs. This approach is demonstrated using a three way predator-prey-food problem where the behavior of the individual should change depending on its energetic situation. The behavior architecture is incrementally evolved.

Genetic approaches for topological active nets optimization

The topological active nets (TANs) model is a deformable model used for image segmentation. It integrates features of region-based and edge-based segmentation techniques so it is able to fit the contours of the objects and model their inner topology. Also, topological changes in its structure allow the detection of concave and convex contours, holes, and several objects in the scene. Since the model deformation is based on the minimization of an energy functional, the adjustment depends on the minimization algorithm. This paper presents two evolutionary approaches to the energy minimization problem in the TAN model. The first proposal is a genetic algorithm with ad hoc operators whereas the second approach is a hybrid model that combines genetic and greedy algorithms. Both evolutionary approaches improve the accuracy of the segmentation even though only the hybrid model allows topological changes in the model structure.

Localisation of the optic disc by means of GA-optimised Topological Active Nets

In this paper we propose a new approach to the optic disc localisation process in digital retinal images by means of Topological Active Nets (TAN). This is a deformable model used for image segmentation that integrates features of region-based and edge-based segmentation techniques, being able to fit the edges of the objects and model their inner topology. In this paper the active nets incorporate new energy terms for the optic disc localisation and their optimisation is performed with a genetic algorithm, with adapted or new ad hoc genetic operators. There is no need of any pre-processing of the images, which allows a quasi automatic localisation of the optic disc. This process also provides a simultaneous segmentation of the disc. We present representative results of optic disc localisations showing the advantages of the approach, with images focusing on the optic disc or on the macula, and with images with different levels of noise and lesion areas.

Differential evolution for protein structure prediction using the HP model

We used Differential Evolution (DE) for the problem of protein structure prediction. We employed the HP model to represent the folding conformations of a protein in a lattice. In this model the nature of amino acids is reduced considering only two types: hydrophobic residues (H) and polar residues (P), which is based on the recognition that hydrophobic interactions are a dominant force in protein folding. Given a primary sequence of amino acids, the problem is to search for the folding structure in the lattice that minimizes an energy potential. This energy reflects the fact that the hydrophobic amino acids have a propensity to form a hydrophobic core. The complexity of the problem has been shown to be NP-hard, with minimal progress achieved in this category of ab initio folding. We combined DE with methods to transform illegal protein conformations to feasible ones, showing the capabilities of the hybridized DE with respect to previous works.

Biologically inspired robot behavior engineering

1. Evolutionary approaches to neural control of rolling, walking, swimming and flying animats or robots.- 2. Behavior coordination and its modification on monkey-type mobile robot.- 3. Visuomotor control in flies and behavior-based agents.- 4. Using evolutionary methods to parameterize neural models: a study of the lamprey central pattern generator.- 5. Biologically inspired neural network approaches to real-time collision-free robot motion planning.- 6. Self-adapting neural networks for mobile robots.- 7. Evolving robots able to integrate sensory-motor information over time.- 8. A non-computationally-intensive neurocontroller for autonomous mobile robot navigation.- 9. Some approaches for reusing behaviour based robot cognitive architectures obtained through evolution.- 10. Modular neural architectures for robotics.- 11. Designing neural control architectures for an autonomous robot using vision to solve complex learning tasks.- 12. Robust estimation of the optical flow based on VQ-BF.- 13. Steps towards one-shot vision-based self-localization.- 14. Computing the optimal trajectory of arm movement: the TOPS (Task Optimization in the Presence of Signal-dependent noise) model.- 15. A general learning approach to visually guided 3D-positioning and pose control of robot arms.

Study of the genetic code adaptability by means of a genetic algorithm

We used simulated evolution to study the adaptability level of the canonical genetic code. An adapted genetic algorithm (GA) searches for optimal hypothetical codes. Adaptability is measured as the average variation of the hydrophobicity that the encoded amino acids undergo when errors or mutations are present in the codons of the hypothetical codes. Different types of mutations and point mutation rates that depend on codon base number are considered in this study. Previous works have used statistical approaches based on randomly generated alternative codes or have used local search techniques to determine an optimum value. In this work, we emphasize what can be concluded from the use of simulated evolution considering the results of previous works. The GA provides more information about the difficulty of the evolution of codes, without contradicting previous studies using statistical or engineering approaches. The GA also shows that, within the coevolution theory, the third base clearly improves the adaptability of the current genetic code.

Considerations in the application of evolution to the generation of robot controllers

Abstract This paper is concerned with different aspects of the use of evolution for the successful generation of real robot Artificial Neural Network (ANN) controllers. Several parameters of an evolutionary/genetic algorithm (GA) and the way they influence the evolution of ANN behavioral controllers for real robots have been contemplated. These parameters include the way the initial populations are distributed, how the individuals are evaluated, the implementation of race schemes, etc. A batch of experiments on the evolution of three types of behaviors with different population sizes have been carried out in order to ascertain their effect on the evolution of the controllers and their validity in real implementations. The results provide a guide to the design of evolutionary algorithms for generating ANN based robot controllers, especially when, due to computational constraints, the populations to be used are small with respect to the complexity of the problem to be solved. The problem of transferring the controllers evolved in simulated environments to the real systems operating in real environments are also considered and we present results of this transference to reality with a robot which has few and extremely noisy sensors.

Topological Active Models optimization with Differential Evolution

The Topological Active Model is an active model focused on segmentation tasks. It provides information about the surfaces and the inside of the detected objects in the scene. The segmentation process turns into a minimization task of the energy functions which control the model deformation. In this work we propose a new optimization method of the segmentation model that uses Differential Evolution as an alternative evolutionary method that minimizes the decisions of the designer with respect to others such as genetic algorithms. Moreover, we hybridized Differential Evolution with a greedy search to integrate the advantages of global and local searches at the same time that the segmentation speed is improved. We also included in the local search the possibility of topological changes to perform a better adjustment in complex surfaces, topological changes that introduce the necessary mechanism to divide the mesh in the case of the presence of several objects in the scene.

Simulated evolution applied to study the genetic code optimality using a model of codon reassignments

BackgroundAs the canonical code is not universal, different theories about its origin and organization have appeared. The optimization or level of adaptation of the canonical genetic code was measured taking into account the harmful consequences resulting from point mutations leading to the replacement of one amino acid for another. There are two basic theories to measure the level of optimization: the statistical approach, which compares the canonical genetic code with many randomly generated alternative ones, and the engineering approach, which compares the canonical code with the best possible alternative.ResultsHere we used a genetic algorithm to search for better adapted hypothetical codes and as a method to guess the difficulty in finding such alternative codes, allowing to clearly situate the canonical code in the fitness landscape. This novel proposal of the use of evolutionary computing provides a new perspective in the open debate between the use of the statistical approach, which postulates that the genetic code conserves amino acid properties far better than expected from a random code, and the engineering approach, which tends to indicate that the canonical genetic code is still far from optimal. We used two models of hypothetical codes: one that reflects the known examples of codon reassignment and the model most used in the two approaches which reflects the current genetic code translation table. Although the standard code is far from a possible optimum considering both models, when the more realistic model of the codon reassignments was used, the evolutionary algorithm had more difficulty to overcome the efficiency of the canonical genetic code.ConclusionsSimulated evolution clearly reveals that the canonical genetic code is far from optimal regarding its optimization. Nevertheless, the efficiency of the canonical code increases when mistranslations are taken into account with the two models, as indicated by the fact that the best possible codes show the patterns of the standard genetic code. Our results are in accordance with the postulates of the engineering approach and indicate that the main arguments of the statistical approach are not enough to its assertion of the extreme efficiency of the canonical genetic code.

Optic Disc Segmentation by Means of GA-Optimized Topological Active Nets

In this paper we propose a new approach to the optic disc segmentation process in digital retinal images by means of Topological Active Nets (TAN). This is a deformable model used for image segmentation that integrates features of region-based and edge-based segmentation techniques, being able to fit the edges of the objects and model their inner topology. The optimization of the Active Nets is performed by a genetic algorithm, with adapted or new ad hoc genetic operators to the problem. The active nets incorporate new energy terms for the optic disc segmentations, without the need of any pre-processing of the images. We present results of optic disc segmentations showing the advantages of the approach.