LUIS FERNANDO Martínez Osorio - Academia.edu (original) (raw)

Papers by LUIS FERNANDO Martínez Osorio

1988 American Control Conference, 1988

ABSTRACT

Resumo: Este artigo apresenta um ambiente virtual tridimensional inteligente e adaptativo para a ... more Resumo: Este artigo apresenta um ambiente virtual tridimensional inteligente e adaptativo para a Educacao a Distância. No ambiente, utilizado para a disponibilizacao de conteudos, a caracteristica de adaptacao esta relacionada com as possibilidades de re-organizacao do mesmo (conforme insercao, remocao ou atualizacao das informacoes) e de personalizacao da apresentacao dos conteudos, conforme interesses e preferencias dos usuarios. Para isto, um perfil de conteudo e um perfil de usuario sao utilizados no processo de adaptacao. Alem disso, o ambiente e habitado por entidades inteligentes que atuam como assistentes dos usuarios durante a navegacao e localizacao de informacoes relevantes, bem como auxiliam na organizacao dos conteudos a serem disponibilizados. Abstract: This article presents an intelligent, adaptable, three-dimensional and virtual environment for distance learning. In the environment, which is used to make contents available, the adaptation characteristic is related wi...

Journal of Intelligent & Fuzzy Systems, 2014

ABSTRACT This work addresses the evolution of an Artificial Neural Network (ANN) to assist in the... more ABSTRACT This work addresses the evolution of an Artificial Neural Network (ANN) to assist in the problem of autonomous navigation of a vehicle in urban environments. We propose a system architecture based on the use of two ANNs, one is trained for image processing, in charge of road recognition and employing template matching. The other ANN is evolved to perform the navigation control. The paper focuses on the evolved ANN, which provides steering and speed control to the autonomous vehicle, corroborating with the Evolutionary Robotic research area and showing its practical viability. The proposed system was tested in several experiments, evaluating not only the impact of different evolutionary computation parameters but also the role of the transfer functions on the evolution of the ANN. Results show that slight variations in the parameters lead to huge differences on the accuracy of the evolutionary process.

El efecto del espacio vital sobre los parámetros productivos y reproductivos fue evaluado en 2325... more El efecto del espacio vital sobre los parámetros productivos y reproductivos fue evaluado en 2325 cuyes de la estación IVITA El Mantaro (300 machos y hembras de recría, 750 machos y hembras de engorde y 200 hembras y 25 machos de primer empadre). El estudio se dividió en siete ensayos evaluando en cada uno de ellos cinco diferentes espacios vitales con cinco diferentes números de animales por poza, totalizando 25 pozas para cada ensayo. Se analizaron variables asociadas a cada etapa productiva (ganancia de peso, consumo de alfalfa, índice de conversión alimenticia, número de cicatrices por peleas, fertilidad, tamaño de camada, mortalidad y el índice beneficio/costo). En los ensayos de recría y engorde de machos los mayores espacios vitales permitieron obtener mayores ganancias de peso, menor consumo de alfalfa, menor índice de conversión alimenticia y menor número de cicatrices por animal; todas estas variables presentaron patrones de respuesta lineal significativos (p = 0,0001 a 0,02 en recría de machos y p = 0,0001 a 0,0007 en engorde de machos). En las hembras de recría se obtuvieron los mismos patrones excepto en la ganancia de peso. En las hembras de engorde la ganancia de peso y el índice beneficio/costo se ajustaron a regresiones cuadráticas alcanzando un óptimo biológico y económico a 0,19 y 0,18 m2/cuy, respectivamente. Se recomiendan los siguientes espacios vitales: 0,16 m2/cuy para machos de recría; 0,14 m2/cuy para hembras de recría; 0,24 m2/cuy para machos de engorde; 0,18 m2/cuy para hembras de engorde y 0,28 m2/cuy para pozas de reproducción.Effect of vital area on productive and reproductive performances was evaluated in 2325 guinea pigs from IVITA El Mantaro Research Station (300 rearing male and female guinea pigs, 750 fattening male and female guinea pigs and 200 female and 25 first mating male guinea pigs). The study was divided in seven assays. Five different vital areas, with five different number of animals per well, were evaluated in each assay. A total of 25 wells were used for each assay. Different variables (weight gain, alfalfa intake, feeding convertion index, number of smears due to fights, fertility, litter size, mortality and profit/cost index) associated to each productive phase were analyzed. Bigger vital areas resulted in greater weight gain, lower alfalfa intake, lower feeding convertion index and lower number of smears due to fights in the assays which involved rearing and fattening males. All of these variables showed significative lineal response patterns (p = 0,0001 to 0,02 for rearing males and p = 0,0001 to 0,0007 in fattening males). The sames patterns were shown for rearing females except for weight gain. For fattening females, weight gain and profit/cost index were adjusted to cuadratic regresion reaching a biological and an echonomic optimus at 0.19 and 0.18 m2/guinea pig. The following vital areas are recommended: 0.16 m2/guinea pig for rearing males; 0.14 m2/guinea pig for rearing females, 0.24 m2/guinea pig for fattening males; 0.18 m2/guinea pig for fattening females, and 0.28 m2/guinea pig for breeding wells.Tesi

The 2006 IEEE International Joint Conference on Neural Network Proceedings, 2006

Abstract—This paper describes the simulation system pro-posed in order to study and to implement ... more Abstract—This paper describes the simulation system pro-posed in order to study and to implement intelligent au-tonomous vehicle control. The developed system can automat-ically drive a vehicle, implementing a robust control system capable of simulating in a realistic way ...

Hybrid Intelligent Systems, 2002

This work presents a new hybrid architecture applied to autonomous mo- bile robot control - HyCAR... more This work presents a new hybrid architecture applied to autonomous mo- bile robot control - HyCAR (Hybrid Control for Autonomous Robots). This archi- tecture provides a robust control for robots as they become able to operate and adapt themselves to different environments and conditions. We designed this new hybrid control architecture, integrating the two main techniques used in robotic control

IFAC Proceedings Volumes, 2013

This paper presents a stereo vision-based autonomous navigation system using a GPS and a modified... more This paper presents a stereo vision-based autonomous navigation system using a GPS and a modified version of the VFH algorithm. In order to obtain a high-accuracy disparity map and meet the time constraints of the real time navigation system, this work proposes the use of a semi-global stereo method. By not suffering the same issues of the regularly used local stereo methods, the employed stereo technique enables the generation of a highly dense, efficient, and accurate disparity map. Obstacles are detected using a method that checks for relative slopes and heights differences. Experimental tests using an electric vehicle in an urban environment were performed to validate the proposed approach.

2014 Joint Conference on Robotics: SBR-LARS Robotics Symposium and Robocontrol, 2014

ABSTRACT This paper presents a Telepresence Mobile Robot using a Kinect sensor as the main percep... more ABSTRACT This paper presents a Telepresence Mobile Robot using a Kinect sensor as the main perception/interface device. Firstly, using the Kinect camera (Webcam) and image processing techniques, it is possible to detect a human face, allowing the robot to track the face, getting closer of a person, moving forward and rotating to get a better position to interact with him/her. Besides that, it is possible to recognize hand gestures using the Kinect 3D sensor, when correctly positioned related to the person. The proposed gesture recognition method tracks the hands positions and movements, when moving it forward towards the robot, and then recognizing a set of predefined gestures/commands. Finally, the 3D perception provided by the Kinect also allows us to detect obstacles, avoiding collisions and helping to move safely and to search for someone in the environment. Practical experiments are presented demonstrating the obtained results in the searching, tracking of faces and obot positioning related to the user, and also, in the gesture recognition.

2010 Latin American Robotics Symposium and Intelligent Robotics Meeting, 2010

... Daniel Sales, Patrick Shinzato, Gustavo Pessin, Denis Wolf and Fernando Osório Mobile Robotic... more ... Daniel Sales, Patrick Shinzato, Gustavo Pessin, Denis Wolf and Fernando Osório Mobile Robotics Laboratory University of São Paulo - USP São Carlos, Brazil dsales@icmc.usp.br, shinzato@icmc.usp.br, pessin@gmail.com, denis@icmc.usp.br, fosorio@gmail.com ...

Communications in Computer and Information Science, 2013

An autonomous ground vehicle has to be able to execute several tasks such as: environment percept... more An autonomous ground vehicle has to be able to execute several tasks such as: environment perception, obstacle detection, and safe navigation. The road shape provides essential information to localization and navigation. It can also be used to identify reference points in the scenario. Crossroads are usual road shapes in urban environments. The detection of these structures is the main focus of this paper. Whereas cameras are sensible to illumination changes, we developed methods that handle LIDAR (Light Detection And Ranging) sensor data to accomplish this task. In the literature, neural networks have not been widely adopted to crossroad detection. One advantage of neural networks is its capability to deal with noisy data, so the detection can be performed even in the presence of other obstacles as cars and pedestrians. Our approach takes advantage of a road detector system that produces curb data and road surface data. Thus we propose a crossroad detector that is performed by an artificial neural network and LIDAR data. We propose two methods (curb detection and road surface detection) for this task. Classification results obtained by different network topologies have been evaluated and the performance compared with ROC graphs. Experimental tests have been carried out to validate the approaches proposed, obtaining good results when compared to other methods in the literature.

2013 III Brazilian Symposium on Computing Systems Engineering, 2013

ABSTRACT Point clouds segmentation is an essential step to improve the performance of obstacle de... more ABSTRACT Point clouds segmentation is an essential step to improve the performance of obstacle detection and classification in areas of autonomous ground vehicles and mobile robotics. This paper presents a study and comparison of the performance of segmentation methods using point clouds coming from a 3D laser sensor, more specifically obtained from a Velodyne HDL32.

2006 Ninth Brazilian Symposium on Neural Networks (SBRN'06), 2006

Page 1. Neural Networks Applied to Gait Control of Physically Based Simulated Robots Milton Rober... more Page 1. Neural Networks Applied to Gait Control of Physically Based Simulated Robots Milton Roberto Heinen and Fernando Santos Osorio Universidade do Vale do Rio dos Sinos (UNISINOS) - Applied Computing mheinen@turing.unisinos.br, fosorio@unisinos.br Abstract ...

Scientia, 2009

Resumo Neste artigo, descreve-se o modelo, a implementação e a avaliação da eficiência de Algorit... more Resumo Neste artigo, descreve-se o modelo, a implementação e a avaliação da eficiência de Algoritmos de Otimização por Enxame de Partículas aplicados à formação e atuação de grupos robóticos. A atuação do grupo robótico é realizada sobre um desastre ambiental do tipo incêndio florestal. São avaliados diversos parâmetros que influenciam o comportamento da otimização, como inércia, confiança, tipos de modelos sociais e tamanho de enxame. Descrevem-se as experiências realizadas, detalhando-se os conjuntos de parâmetros que permitem obter resultados positivos e também negativos. Os resultados das simulações demonstram que, com um conjunto adequado de parâmetros, é possível obter posições satisfatórias para atuação do grupo robótico.

IFAC Proceedings Volumes, 2010

This paper presents the modeling, implementation and evaluation of the Particle Swarm Optimizatio... more This paper presents the modeling, implementation and evaluation of the Particle Swarm Optimization (PSO) applied to intelligent vehicles group formation and coordination. The robotic task discussed in this paper is performed over a natural disaster scenario, simulated as a forest fire. The intelligent vehicles squad mission should surround the fire and avoid fire's propagation. Experiments have been carried out with several PSO parameter's variation (e.g. inertia, confidence, social models, swarm size) seeking to get the more efficient optimization for the formation of the group. This paper describes all performed experiments detailing all sets of parameters, including positive and negative results. The simulation's results showed that with an adequate set of parameters is possible to get satisfactory strategic positions for a multirobotic system's operation using PSO.

16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013), 2013

ABSTRACT Autonomous navigation in agricultural environments is a promising research topic for rob... more ABSTRACT Autonomous navigation in agricultural environments is a promising research topic for robotics, with several practical applications. This paper presents an obstacle detection system to operate in field scenarios that can accurately discern high and low vegetation from other types of obstacles. Our algorithm is composed by three steps: (i) obstacle detection based on geometric information; (ii) clustering of detected obstacles; and (iii) filtering false positive detections using Bayesian classifiers. Several experimental tests have been carried out in citrus plantations. The results showed that our approach is able to correctly identify obstacles, classifying them as people, bushes, animals, and grass of different heights. In addition, the proposed approach could also be employed as a general framework for stereo-based obstacle detection.

2012 Second Brazilian Conference on Critical Embedded Systems, 2012

This paper demonstrates a method for global localization of autonomous mobile robots based on the... more This paper demonstrates a method for global localization of autonomous mobile robots based on the creation of visual memory maps, through detection and description of reference points from captured images, associated to odometer data in a specific environment. The proposed procedure, coupled with specific knowledge of the environment, allows for localization to be achieved through the pairing of these memorized

Electronics, Robotics and Automotive Mechanics Conference (CERMA 2007), 2007

Page 1. Applying Genetic Algorithms to Control Gait of Simulated Robots Milton Roberto Heinen UFR... more Page 1. Applying Genetic Algorithms to Control Gait of Simulated Robots Milton Roberto Heinen UFRGS – Informatics Institute CEP 91501-970, Porto Alegre, Brazil mrheinen@inf.ufrgs.br Fernando Santos Osório UNISINOS ...

2014 IEEE Intelligent Vehicles Symposium Proceedings, 2014

Robust and stable control is a requirement for navigation of self-driving cars. Some approaches i... more Robust and stable control is a requirement for navigation of self-driving cars. Some approaches in the literature depend on a high number of parameters that are often difficult to estimate. A poor selection of these parameters often reduces considerably the efficiency of the control algorithms. In this paper we propose a simplified control system for autonomous vehicles that depends on a reduced number of parameters that can be easily set. This control system is composed of longitudinal and lateral controllers. The longitudinal controller is responsible for regulating the vehicle's cruise velocity while the lateral controller steers the vehicle's wheels for path tracking. Simulated and experimental tests have been carried out with the CaRINA II platform in the university campus with positive results.

Brazilian Symposium on Neural Networks, 2002

This paper presents the SEVA system (Autonomous Vehicle Parking Simulator). This tool implements ... more This paper presents the SEVA system (Autonomous Vehicle Parking Simulator). This tool implements a robust control system for autonomous vehicle parking based on a FSA (Finite-State Automata) and also based on FSA ob- tained from trained J-CC ANNs (Jordan Cascade- Correlation Artificial Neural Networks).

1988 American Control Conference, 1988

ABSTRACT

Resumo: Este artigo apresenta um ambiente virtual tridimensional inteligente e adaptativo para a ... more Resumo: Este artigo apresenta um ambiente virtual tridimensional inteligente e adaptativo para a Educacao a Distância. No ambiente, utilizado para a disponibilizacao de conteudos, a caracteristica de adaptacao esta relacionada com as possibilidades de re-organizacao do mesmo (conforme insercao, remocao ou atualizacao das informacoes) e de personalizacao da apresentacao dos conteudos, conforme interesses e preferencias dos usuarios. Para isto, um perfil de conteudo e um perfil de usuario sao utilizados no processo de adaptacao. Alem disso, o ambiente e habitado por entidades inteligentes que atuam como assistentes dos usuarios durante a navegacao e localizacao de informacoes relevantes, bem como auxiliam na organizacao dos conteudos a serem disponibilizados. Abstract: This article presents an intelligent, adaptable, three-dimensional and virtual environment for distance learning. In the environment, which is used to make contents available, the adaptation characteristic is related wi...

Journal of Intelligent & Fuzzy Systems, 2014

ABSTRACT This work addresses the evolution of an Artificial Neural Network (ANN) to assist in the... more ABSTRACT This work addresses the evolution of an Artificial Neural Network (ANN) to assist in the problem of autonomous navigation of a vehicle in urban environments. We propose a system architecture based on the use of two ANNs, one is trained for image processing, in charge of road recognition and employing template matching. The other ANN is evolved to perform the navigation control. The paper focuses on the evolved ANN, which provides steering and speed control to the autonomous vehicle, corroborating with the Evolutionary Robotic research area and showing its practical viability. The proposed system was tested in several experiments, evaluating not only the impact of different evolutionary computation parameters but also the role of the transfer functions on the evolution of the ANN. Results show that slight variations in the parameters lead to huge differences on the accuracy of the evolutionary process.

El efecto del espacio vital sobre los parámetros productivos y reproductivos fue evaluado en 2325... more El efecto del espacio vital sobre los parámetros productivos y reproductivos fue evaluado en 2325 cuyes de la estación IVITA El Mantaro (300 machos y hembras de recría, 750 machos y hembras de engorde y 200 hembras y 25 machos de primer empadre). El estudio se dividió en siete ensayos evaluando en cada uno de ellos cinco diferentes espacios vitales con cinco diferentes números de animales por poza, totalizando 25 pozas para cada ensayo. Se analizaron variables asociadas a cada etapa productiva (ganancia de peso, consumo de alfalfa, índice de conversión alimenticia, número de cicatrices por peleas, fertilidad, tamaño de camada, mortalidad y el índice beneficio/costo). En los ensayos de recría y engorde de machos los mayores espacios vitales permitieron obtener mayores ganancias de peso, menor consumo de alfalfa, menor índice de conversión alimenticia y menor número de cicatrices por animal; todas estas variables presentaron patrones de respuesta lineal significativos (p = 0,0001 a 0,02 en recría de machos y p = 0,0001 a 0,0007 en engorde de machos). En las hembras de recría se obtuvieron los mismos patrones excepto en la ganancia de peso. En las hembras de engorde la ganancia de peso y el índice beneficio/costo se ajustaron a regresiones cuadráticas alcanzando un óptimo biológico y económico a 0,19 y 0,18 m2/cuy, respectivamente. Se recomiendan los siguientes espacios vitales: 0,16 m2/cuy para machos de recría; 0,14 m2/cuy para hembras de recría; 0,24 m2/cuy para machos de engorde; 0,18 m2/cuy para hembras de engorde y 0,28 m2/cuy para pozas de reproducción.Effect of vital area on productive and reproductive performances was evaluated in 2325 guinea pigs from IVITA El Mantaro Research Station (300 rearing male and female guinea pigs, 750 fattening male and female guinea pigs and 200 female and 25 first mating male guinea pigs). The study was divided in seven assays. Five different vital areas, with five different number of animals per well, were evaluated in each assay. A total of 25 wells were used for each assay. Different variables (weight gain, alfalfa intake, feeding convertion index, number of smears due to fights, fertility, litter size, mortality and profit/cost index) associated to each productive phase were analyzed. Bigger vital areas resulted in greater weight gain, lower alfalfa intake, lower feeding convertion index and lower number of smears due to fights in the assays which involved rearing and fattening males. All of these variables showed significative lineal response patterns (p = 0,0001 to 0,02 for rearing males and p = 0,0001 to 0,0007 in fattening males). The sames patterns were shown for rearing females except for weight gain. For fattening females, weight gain and profit/cost index were adjusted to cuadratic regresion reaching a biological and an echonomic optimus at 0.19 and 0.18 m2/guinea pig. The following vital areas are recommended: 0.16 m2/guinea pig for rearing males; 0.14 m2/guinea pig for rearing females, 0.24 m2/guinea pig for fattening males; 0.18 m2/guinea pig for fattening females, and 0.28 m2/guinea pig for breeding wells.Tesi

The 2006 IEEE International Joint Conference on Neural Network Proceedings, 2006

Abstract—This paper describes the simulation system pro-posed in order to study and to implement ... more Abstract—This paper describes the simulation system pro-posed in order to study and to implement intelligent au-tonomous vehicle control. The developed system can automat-ically drive a vehicle, implementing a robust control system capable of simulating in a realistic way ...

Hybrid Intelligent Systems, 2002

This work presents a new hybrid architecture applied to autonomous mo- bile robot control - HyCAR... more This work presents a new hybrid architecture applied to autonomous mo- bile robot control - HyCAR (Hybrid Control for Autonomous Robots). This archi- tecture provides a robust control for robots as they become able to operate and adapt themselves to different environments and conditions. We designed this new hybrid control architecture, integrating the two main techniques used in robotic control

IFAC Proceedings Volumes, 2013

This paper presents a stereo vision-based autonomous navigation system using a GPS and a modified... more This paper presents a stereo vision-based autonomous navigation system using a GPS and a modified version of the VFH algorithm. In order to obtain a high-accuracy disparity map and meet the time constraints of the real time navigation system, this work proposes the use of a semi-global stereo method. By not suffering the same issues of the regularly used local stereo methods, the employed stereo technique enables the generation of a highly dense, efficient, and accurate disparity map. Obstacles are detected using a method that checks for relative slopes and heights differences. Experimental tests using an electric vehicle in an urban environment were performed to validate the proposed approach.

2014 Joint Conference on Robotics: SBR-LARS Robotics Symposium and Robocontrol, 2014

ABSTRACT This paper presents a Telepresence Mobile Robot using a Kinect sensor as the main percep... more ABSTRACT This paper presents a Telepresence Mobile Robot using a Kinect sensor as the main perception/interface device. Firstly, using the Kinect camera (Webcam) and image processing techniques, it is possible to detect a human face, allowing the robot to track the face, getting closer of a person, moving forward and rotating to get a better position to interact with him/her. Besides that, it is possible to recognize hand gestures using the Kinect 3D sensor, when correctly positioned related to the person. The proposed gesture recognition method tracks the hands positions and movements, when moving it forward towards the robot, and then recognizing a set of predefined gestures/commands. Finally, the 3D perception provided by the Kinect also allows us to detect obstacles, avoiding collisions and helping to move safely and to search for someone in the environment. Practical experiments are presented demonstrating the obtained results in the searching, tracking of faces and obot positioning related to the user, and also, in the gesture recognition.

2010 Latin American Robotics Symposium and Intelligent Robotics Meeting, 2010

... Daniel Sales, Patrick Shinzato, Gustavo Pessin, Denis Wolf and Fernando Osório Mobile Robotic... more ... Daniel Sales, Patrick Shinzato, Gustavo Pessin, Denis Wolf and Fernando Osório Mobile Robotics Laboratory University of São Paulo - USP São Carlos, Brazil dsales@icmc.usp.br, shinzato@icmc.usp.br, pessin@gmail.com, denis@icmc.usp.br, fosorio@gmail.com ...

Communications in Computer and Information Science, 2013

An autonomous ground vehicle has to be able to execute several tasks such as: environment percept... more An autonomous ground vehicle has to be able to execute several tasks such as: environment perception, obstacle detection, and safe navigation. The road shape provides essential information to localization and navigation. It can also be used to identify reference points in the scenario. Crossroads are usual road shapes in urban environments. The detection of these structures is the main focus of this paper. Whereas cameras are sensible to illumination changes, we developed methods that handle LIDAR (Light Detection And Ranging) sensor data to accomplish this task. In the literature, neural networks have not been widely adopted to crossroad detection. One advantage of neural networks is its capability to deal with noisy data, so the detection can be performed even in the presence of other obstacles as cars and pedestrians. Our approach takes advantage of a road detector system that produces curb data and road surface data. Thus we propose a crossroad detector that is performed by an artificial neural network and LIDAR data. We propose two methods (curb detection and road surface detection) for this task. Classification results obtained by different network topologies have been evaluated and the performance compared with ROC graphs. Experimental tests have been carried out to validate the approaches proposed, obtaining good results when compared to other methods in the literature.

2013 III Brazilian Symposium on Computing Systems Engineering, 2013

ABSTRACT Point clouds segmentation is an essential step to improve the performance of obstacle de... more ABSTRACT Point clouds segmentation is an essential step to improve the performance of obstacle detection and classification in areas of autonomous ground vehicles and mobile robotics. This paper presents a study and comparison of the performance of segmentation methods using point clouds coming from a 3D laser sensor, more specifically obtained from a Velodyne HDL32.

2006 Ninth Brazilian Symposium on Neural Networks (SBRN'06), 2006

Page 1. Neural Networks Applied to Gait Control of Physically Based Simulated Robots Milton Rober... more Page 1. Neural Networks Applied to Gait Control of Physically Based Simulated Robots Milton Roberto Heinen and Fernando Santos Osorio Universidade do Vale do Rio dos Sinos (UNISINOS) - Applied Computing mheinen@turing.unisinos.br, fosorio@unisinos.br Abstract ...

Scientia, 2009

Resumo Neste artigo, descreve-se o modelo, a implementação e a avaliação da eficiência de Algorit... more Resumo Neste artigo, descreve-se o modelo, a implementação e a avaliação da eficiência de Algoritmos de Otimização por Enxame de Partículas aplicados à formação e atuação de grupos robóticos. A atuação do grupo robótico é realizada sobre um desastre ambiental do tipo incêndio florestal. São avaliados diversos parâmetros que influenciam o comportamento da otimização, como inércia, confiança, tipos de modelos sociais e tamanho de enxame. Descrevem-se as experiências realizadas, detalhando-se os conjuntos de parâmetros que permitem obter resultados positivos e também negativos. Os resultados das simulações demonstram que, com um conjunto adequado de parâmetros, é possível obter posições satisfatórias para atuação do grupo robótico.

IFAC Proceedings Volumes, 2010

This paper presents the modeling, implementation and evaluation of the Particle Swarm Optimizatio... more This paper presents the modeling, implementation and evaluation of the Particle Swarm Optimization (PSO) applied to intelligent vehicles group formation and coordination. The robotic task discussed in this paper is performed over a natural disaster scenario, simulated as a forest fire. The intelligent vehicles squad mission should surround the fire and avoid fire's propagation. Experiments have been carried out with several PSO parameter's variation (e.g. inertia, confidence, social models, swarm size) seeking to get the more efficient optimization for the formation of the group. This paper describes all performed experiments detailing all sets of parameters, including positive and negative results. The simulation's results showed that with an adequate set of parameters is possible to get satisfactory strategic positions for a multirobotic system's operation using PSO.

16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013), 2013

ABSTRACT Autonomous navigation in agricultural environments is a promising research topic for rob... more ABSTRACT Autonomous navigation in agricultural environments is a promising research topic for robotics, with several practical applications. This paper presents an obstacle detection system to operate in field scenarios that can accurately discern high and low vegetation from other types of obstacles. Our algorithm is composed by three steps: (i) obstacle detection based on geometric information; (ii) clustering of detected obstacles; and (iii) filtering false positive detections using Bayesian classifiers. Several experimental tests have been carried out in citrus plantations. The results showed that our approach is able to correctly identify obstacles, classifying them as people, bushes, animals, and grass of different heights. In addition, the proposed approach could also be employed as a general framework for stereo-based obstacle detection.

2012 Second Brazilian Conference on Critical Embedded Systems, 2012

This paper demonstrates a method for global localization of autonomous mobile robots based on the... more This paper demonstrates a method for global localization of autonomous mobile robots based on the creation of visual memory maps, through detection and description of reference points from captured images, associated to odometer data in a specific environment. The proposed procedure, coupled with specific knowledge of the environment, allows for localization to be achieved through the pairing of these memorized

Electronics, Robotics and Automotive Mechanics Conference (CERMA 2007), 2007

Page 1. Applying Genetic Algorithms to Control Gait of Simulated Robots Milton Roberto Heinen UFR... more Page 1. Applying Genetic Algorithms to Control Gait of Simulated Robots Milton Roberto Heinen UFRGS – Informatics Institute CEP 91501-970, Porto Alegre, Brazil mrheinen@inf.ufrgs.br Fernando Santos Osório UNISINOS ...

2014 IEEE Intelligent Vehicles Symposium Proceedings, 2014

Robust and stable control is a requirement for navigation of self-driving cars. Some approaches i... more Robust and stable control is a requirement for navigation of self-driving cars. Some approaches in the literature depend on a high number of parameters that are often difficult to estimate. A poor selection of these parameters often reduces considerably the efficiency of the control algorithms. In this paper we propose a simplified control system for autonomous vehicles that depends on a reduced number of parameters that can be easily set. This control system is composed of longitudinal and lateral controllers. The longitudinal controller is responsible for regulating the vehicle's cruise velocity while the lateral controller steers the vehicle's wheels for path tracking. Simulated and experimental tests have been carried out with the CaRINA II platform in the university campus with positive results.

Brazilian Symposium on Neural Networks, 2002

This paper presents the SEVA system (Autonomous Vehicle Parking Simulator). This tool implements ... more This paper presents the SEVA system (Autonomous Vehicle Parking Simulator). This tool implements a robust control system for autonomous vehicle parking based on a FSA (Finite-State Automata) and also based on FSA ob- tained from trained J-CC ANNs (Jordan Cascade- Correlation Artificial Neural Networks).