Search

This article describes two vision-based behaviors designed for an autonomous legged robot. These behaviors have been designed in a modular way in order to be able to integrate them in an architecture named DSH (Dynamic Schema Hierarchies), which is also briefly described. These behaviors have been tested in office indoor environments and experiments carried out are also described in this paper. The platform used in these experiments carried out are also described in theis paper. The platform used in these experiments has been the Sony AIBO robot. The developed behaviors provide the robot with the follow wall and follow person capabilities, using as main sensor the robot on-board camera

Este artículo presenta un ejemplo de cooperación borrosa entre un grupo de robots como una solución alternativa a los protocolos clásicos utilizados en las arquitecturas de cooperación entre robots. Para ello se ha procedido a integrar trabajos previos realizados en el Laboratorio de Agentes Inteligentes (LAI) sobre comportamientos borrosos y sobre arquitecturas de cooperación. Se han considerado tres tipos diferentes de protocolos borrosos y se han probado tanto en un simulador como sobre un robot real. Por último, se presenta la descripción de los experimentos realizados así como los resultados obtenidos

P. 475-483 This paper presents an architecture for the control of autonomus agents that allows explicit cooperation among them. The structure of the software agents controlling the robots is based an a general purpose multi-agent architecture based on a two level approach. One level is composed of reactive skills capable of archieving simple actions by their own. The other is based an an agenda used as an opportunistic planning mechanism to compound, activate and coordinate the basic skills. This agenda handles actions both from the internal goals of the robot or from other robots. This paper describes the work already accomplished, as well as the issues arising from the implementation of the architecture and its use the RoboCup domain SI

A new methodological approach to solve the problem of integrating different behaviors in autonomous robots is developed in this paper. It relies on the succesful integration of several fuzzy relational algorithms (Zadeh73), each one specialized in a different task. The whole system is based on a continiously act on sensor data (reactive) approach. The problem is how to got an intelligent behavior from low level reactive controllers. Our assumption is that we can obtain adecuate response to unkown external situations from the fusion of basic actions

This paper is concerned with the learning of basic behaviors in autonomous robots. In this way, we present a method for the adaptation of basic reactive behaviors implemented as fuzzy controllers applying a genetic algorithm to the evolution of the fuzzy rule system. In this sense, we show our experiments in the evolution of control rules based on symbolic concepts represented as linguistic labels. The rules will be formulated in a fuzzy way and in order to test the rules obtained in each generation of the genetic algorithm a real robot has been used. The individual with the best performance is chosen to generate a new population: the elite strategy. All the new individuals were tested in the same real environment. In conclusion, the individuals of the last generation offer a set of rules that provides better performance than the ones designed by a non-expert designer

The main goal of the project described in this paper is to create a security system using autonomous surveillance robots that use SODAR-like detection system sensors, working with acoustic signals in air environment and navigation base on Geographic Information System and Markov's models. The surveillance system based on SONAR provides great information from the environment, even lets you see behind objects (rebounds effects) whose manipulation offers a great added value to surveillance The guide system will implement in one hand a local navigation module directed to avoid obstacles based on classical techniques and using the new SODAR sensor. On the other hand a global navigation module will be implemented using preset trajectories and gradient techniques and an auto-location system. One of the greatest challenges obtained is the definition of the VIGICOPVar variable that defines, depending on the environment and safety parameters, the probability of intrusion. Surveillance experts of GRUPO NORTE (multinational company with security expertise of more than 38 years) have worked In the definition and validation of the model. The monitoring robots will be controlled in a centralized way from an alarm center from where you can manage all information relating to intrusion detected. VIGICOP is the low cost surveillance robot which provides new/full information interactive surveillance information

This article describes two vision-based behaviors designed for an autonomous robot. These behaviors have been designed and implemented using schemas according to an architecture named DSH (Dynamic Schema Hierarchies), which is also briefly described. The reactive behaviors implemented have been tested in our robotic soccer players, running in EyeBot commercial platforms. The developed behaviors provide the robot with the follow ball, and the follow border capabilities, using only the local camera and the robot on-board low processing power. The design of the control schemas, the algorithms used to analyze the images in perceptual schemas and relevant implementation issues are also described in this paper

P. 33-41 This paper concerns the learning of basic behaviors in an autonomous robot. It presents a method to adapt basic reactive behaviors using a genetic algorithm. Behaviors are implemented as fuzzy controllers and the genetic algorithm is used to evolve their rules. These rules will be formulated in a fuzzy way using prefixed linguistic labels. In order to test the rules obtained in each generation of the genetic evolution process, a real robot has been used. Numerical results from the evolution rate of the different experiments, as well as an example of the fuzzy rules obtained, are presented and discussed SI

P. 68-74 Este artículo se enmarca dentro del campo de las arquitecturas de cooperación entre robots. En concreto, el trabajo presenta un estudio comparativo del rendimiento de tres protocolos de comunicación borrosos entre dos robots autónomos

In this paper a fuzzy interaction among robots in a group is presented as an alternative solution to the classical protocols used in architectures for the cooperation of robots. In order to achieve this goal, the integration of previous works carried out on fuzzy behaviors and on cooperative architectures at the Intelligent Agents Lab. (LAI) has been used. This cooperation/coordination protocol is necessary to successfully control a group of autonomous robots. The protocol will take into account the fuzzy controllers used in the design of the robots. Three different fuzzy protocols have been considered and tested, both in a simulator and in real robots. The results of the experiments carried out and the conclusions obtained are presented.