Marcio Lobo Netto
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Marcio Lobo Netto has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Joao Francisco Justo
Escola Politcnica, Universidade de So Paulo, So Paulo, Brazil
4 shared publications - Caio Augustus Morais Bolzani
Department of Electronic Systems, University of São Paulo, Av. Prof. Luciano Gualberto, tr. 3, 158, São Paulo, Brazil
2 shared publications - Joao Ranhel
CECS – Engenharia de Informação, UFABC – Univ. Federal do ABC, Santo André, Brazil
2 shared publications José Roberto C. Piqueira
Escola Politécnica da Universidade de São Paulo, Avenida Prof. Luciano Gualberto, Travessa 3, No. 158, 05508-900 São Paulo, SP, Brazil
2 shared publicationsJoão Eduardo Kögler
2 shared publications- Max Mauro Dias Santos
Department of Electronics, Federal University of Technology - Parana, Ponta Grossa, Brazil
2 shared publications
Latest Publications
- Open Access
Multi-Agent Sensor Fusion Methodology Using Deep Reinforcement Learning: Vehicle Sensors to Localization
Despite recent major advances in autonomous driving, several challenges remain. Even with modern advanced sensors and processing systems, vehicles are still unable to detect all possible obstacles present in complex urban settings and under diverse environmental conditions. Consequently, numerous studies have investigated artificial intelligence methods to improve vehicle perception capabilities. This paper presents a new methodology using a framework named CarAware, which fuses multiple types of sensor data to predict vehicle positions using Deep Reinforcement Learning (DRL). Unlike traditional DRL applications centered on control, this approach focuses on perception. As a case study, the PPO algorithm was used to train and evaluate the effectiveness of this methodology.
Open AccessMulti-Agent Sensor Fusion Methodology Using Deep Reinforcement Learning: Vehicle Sensors to Localization
Despite recent major advances in autonomous driving, several challenges remain. Even with modern advanced sensors and processing systems, vehicles are still unable to detect all possible obstacles present in complex urban settings and under diverse environmental conditions. Consequently, numerous studies have investigated artificial intelligence methods to improve vehicle perception capabilities. This paper presents a new methodology using a framework named CarAware, which fuses multiple types of sensor data to predict vehicle positions using Deep Reinforcement Learning (DRL). Unlike traditional DRL applications centered on control, this approach focuses on perception. As a case study, the PPO algorithm was used to train and evaluate the effectiveness of this methodology.
- Open Access
- 1 December 2023
- Published by FapUNIFESP (SciELO) in Estudos Avançados
- Vol. 37 (109) , 345-360
Cidades cognitivas: Utopia tecnológica ou revolução urbana?
Os avanços tecnológicos das últimas duas décadas, particularmente nos dispositivos eletrônicos e computação, abriram o caminho para uma extensa lista de potenciais aplicações, em diversas áreas, impactando decisivamente as ações humanas. A proliferação dos equipamentos eletrônicos, tanto de uso pessoal como de coletivo, e conectados de forma ininterrupta, favoreceu a criação e disponibilização de um extenso conjunto de informações, sobre demandas individuais e coletivas, podendo ser utilizadas para gerenciar uma estrutura urbana de forma mais eficiente e otimizada. Surge assim o conceito de estruturas urbanas com uma forma de inteligência subliminar, as cidades inteligentes. Um passo tecnológico complementar foi a proliferação dos gadgets e apps, permitindo um crescimento sem precedentes no volume de dados disponíveis, e que podem ser usados por algoritmos de inteligência artificial para tomar decisões autônomas em tempo real na gestão da infraestrutura urbana, as cidades cognitivas. Embora isso possa representar uma grande revolução urbana, com a melhoria na qualidade de vida nas cidades, o funcionamento de uma estrutura coletiva de forma totalmente autônoma traz consigo profundos questionamentos éticos e morais. Aqui, discutimos as tecnologias que levaram a essa revolução urbana, e os consequentes desafios da humanidade para estabelecer os parâmetros limitantes para o uso dessas tecnologias, e de como este novo cenário urbano resgata o debate sobre o papel da tecnologia na sociedade e no progresso da humanidade.
- 1 December 2023
- Published by FapUNIFESP (SciELO) in Estudos Avançados
- Vol. 37 (109) , 345-360
Open AccessCidades cognitivas: Utopia tecnológica ou revolução urbana?
Os avanços tecnológicos das últimas duas décadas, particularmente nos dispositivos eletrônicos e computação, abriram o caminho para uma extensa lista de potenciais aplicações, em diversas áreas, impactando decisivamente as ações humanas. A proliferação dos equipamentos eletrônicos, tanto de uso pessoal como de coletivo, e conectados de forma ininterrupta, favoreceu a criação e disponibilização de um extenso conjunto de informações, sobre demandas individuais e coletivas, podendo ser utilizadas para gerenciar uma estrutura urbana de forma mais eficiente e otimizada. Surge assim o conceito de estruturas urbanas com uma forma de inteligência subliminar, as cidades inteligentes. Um passo tecnológico complementar foi a proliferação dos gadgets e apps, permitindo um crescimento sem precedentes no volume de dados disponíveis, e que podem ser usados por algoritmos de inteligência artificial para tomar decisões autônomas em tempo real na gestão da infraestrutura urbana, as cidades cognitivas. Embora isso possa representar uma grande revolução urbana, com a melhoria na qualidade de vida nas cidades, o funcionamento de uma estrutura coletiva de forma totalmente autônoma traz consigo profundos questionamentos éticos e morais. Aqui, discutimos as tecnologias que levaram a essa revolução urbana, e os consequentes desafios da humanidade para estabelecer os parâmetros limitantes para o uso dessas tecnologias, e de como este novo cenário urbano resgata o debate sobre o papel da tecnologia na sociedade e no progresso da humanidade.
- Open Access
Automatic Calibration of Microscopic Traffic Simulation Models Using Artificial Neural Networks
Traffic simulations are valuable tools for urban mobility planning and operation, particularly in large cities. Simulation-based microscopic models have enabled traffic engineers to understand local transit and transport behaviors more deeply and manage urban mobility. However, for the simulations to be effective, the transport network and user behavior parameters must be calibrated to mirror real scenarios. In general, calibration is performed manually by traffic engineers who use their knowledge and experience to adjust the parameters of the simulator. Unfortunately, there is still no systematic and automatic process for calibrating traffic simulation networks, although some methods have been proposed in the literature. This study proposes a methodology that facilitates the calibration process, where an artificial neural network (ANN) is trained to learn the behavior of the transport network of interest. The ANN used is the Multi-Layer Perceptron (MLP), trained with back-propagation methods. Based on this learning procedure, the neural network can select the optimized values of the simulation parameters that best mimic the traffic conditions of interest. Experiments considered two microscopic models of traffic and two psychophysical models (Wiedemann 74 and Wiedemann 99). The microscopic traffic models are located in the metropolitan region of São Paulo, Brazil. Moreover, we tested the different configurations of the MLP (layers and numbers of neurons) as well as several variations of the backpropagation training method: Stochastic Gradient Descent (SGD), Adam, Adagrad, Adadelta, Adamax, and Nadam. The results of the experiments show that the proposed methodology is accurate and efficient, leading to calibration with a correlation coefficient greater than 0.8, when the calibrated parameters generate more visible effects on the road network, such as travel times, vehicle counts, and average speeds. For the psychophysical parameters, in the most simplified model (W74), the correlation coefficient was greater than 0.7. The advantage of using ANN for the automatic calibration of simulation parameters is that it allows traffic engineers to carry out comprehensive studies on a large number of future scenarios, such as at different times of the day, as well as on different days of the week and months of the year.
Open AccessAutomatic Calibration of Microscopic Traffic Simulation Models Using Artificial Neural Networks
Traffic simulations are valuable tools for urban mobility planning and operation, particularly in large cities. Simulation-based microscopic models have enabled traffic engineers to understand local transit and transport behaviors more deeply and manage urban mobility. However, for the simulations to be effective, the transport network and user behavior parameters must be calibrated to mirror real scenarios. In general, calibration is performed manually by traffic engineers who use their knowledge and experience to adjust the parameters of the simulator. Unfortunately, there is still no systematic and automatic process for calibrating traffic simulation networks, although some methods have been proposed in the literature. This study proposes a methodology that facilitates the calibration process, where an artificial neural network (ANN) is trained to learn the behavior of the transport network of interest. The ANN used is the Multi-Layer Perceptron (MLP), trained with back-propagation methods. Based on this learning procedure, the neural network can select the optimized values of the simulation parameters that best mimic the traffic conditions of interest. Experiments considered two microscopic models of traffic and two psychophysical models (Wiedemann 74 and Wiedemann 99). The microscopic traffic models are located in the metropolitan region of São Paulo, Brazil. Moreover, we tested the different configurations of the MLP (layers and numbers of neurons) as well as several variations of the backpropagation training method: Stochastic Gradient Descent (SGD), Adam, Adagrad, Adadelta, Adamax, and Nadam. The results of the experiments show that the proposed methodology is accurate and efficient, leading to calibration with a correlation coefficient greater than 0.8, when the calibrated parameters generate more visible effects on the road network, such as travel times, vehicle counts, and average speeds. For the psychophysical parameters, in the most simplified model (W74), the correlation coefficient was greater than 0.7. The advantage of using ANN for the automatic calibration of simulation parameters is that it allows traffic engineers to carry out comprehensive studies on a large number of future scenarios, such as at different times of the day, as well as on different days of the week and months of the year.
- 1 August 2020
- Published by Elsevier in Vehicular Communications
- Vol. 24
An efficient formulation for optimization of FlexRay frame scheduling
- 1 August 2020
- Published by Elsevier in Vehicular Communications
- Vol. 24
An efficient formulation for optimization of FlexRay frame scheduling
- Open Access
- 22 July 2019
- Published by Springer Nature in EURASIP Journal on Advances in Signal Processing
- Vol. 2019 (1) , 1-11
Designing digital filter banks using wavelets
- 22 July 2019
- Published by Springer Nature in EURASIP Journal on Advances in Signal Processing
- Vol. 2019 (1) , 1-11
Open AccessDesigning digital filter banks using wavelets
- 1 August 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15301834,p. 33-40
A Comparison of Global Illumination Methods Using Perceptual Quality Metrics
This work aims to build a comparison basis for analyzing global illumination methods. We have compared six state-of-the-art global illumination methods, ranging from Monte Carlo Path Tracing techniques to Density Estimation methods such as Progressive Photon Mapping, and the mixture approach Vertex Connection and Merging, using nine test scenes with very different characteristics, including many different light types, illumination conditions and BRDFs, exploring many different light scattering events. In order to compare results, the perceptual quality metrics SSIM and HDR-VDP-2 were used. We provide a complete set of convergence rate curves and results for all test scenes and all analyzed methods. We also discuss strategies to generate reference images for each case. We concluded that in general cases the overhead introduced by BPT and VCM is well compensated by the quality of the produced images, however VCM can handle more interesting effects. We also showed there are usual cases such as interiors with strong indirect illumination in which Light Tracing method excels.
- 1 August 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15301834,p. 33-40
A Comparison of Global Illumination Methods Using Perceptual Quality Metrics
This work aims to build a comparison basis for analyzing global illumination methods. We have compared six state-of-the-art global illumination methods, ranging from Monte Carlo Path Tracing techniques to Density Estimation methods such as Progressive Photon Mapping, and the mixture approach Vertex Connection and Merging, using nine test scenes with very different characteristics, including many different light types, illumination conditions and BRDFs, exploring many different light scattering events. In order to compare results, the perceptual quality metrics SSIM and HDR-VDP-2 were used. We provide a complete set of convergence rate curves and results for all test scenes and all analyzed methods. We also discuss strategies to generate reference images for each case. We concluded that in general cases the overhead introduced by BPT and VCM is well compensated by the quality of the produced images, however VCM can handle more interesting effects. We also showed there are usual cases such as interiors with strong indirect illumination in which Light Tracing method excels.
- 1 December 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 120-125
A system for social network analysis
As new technologies advance, simulations and computational experiments are increasingly interesting alternatives for the scientific study of specific working principles of a real system. The goal of this project is the development of a system to support experiments in primates behavior psychology that seeks evaluating the learning processes (usage of tools for breaking encapsulated fruits) through observation and experiments that may emerge in an autonomous agents group (represented by Capuchin monkeys). The experiment focuses on analyzing data through simulations for the prediction of social behavior associated to some cognitive skills. The dissemination of information occur through the group social structure, which is represented by a graph object that has been built based on theorems and algorithms from the graph theory. When a simulation of a complex society behavior is performed, it is important to highlight that it takes places over a mathematical model which is not able to represent neither animals mental properties nor the real shape of a society behavior. Thus, the behavior of these agents is naturally limited to characteristics present in computational models, so agents mental models and their corresponding social behaviors are naturally reality sim plifications.
- 1 December 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 120-125
A system for social network analysis
As new technologies advance, simulations and computational experiments are increasingly interesting alternatives for the scientific study of specific working principles of a real system. The goal of this project is the development of a system to support experiments in primates behavior psychology that seeks evaluating the learning processes (usage of tools for breaking encapsulated fruits) through observation and experiments that may emerge in an autonomous agents group (represented by Capuchin monkeys). The experiment focuses on analyzing data through simulations for the prediction of social behavior associated to some cognitive skills. The dissemination of information occur through the group social structure, which is represented by a graph object that has been built based on theorems and algorithms from the graph theory. When a simulation of a complex society behavior is performed, it is important to highlight that it takes places over a mathematical model which is not able to represent neither animals mental properties nor the real shape of a society behavior. Thus, the behavior of these agents is naturally limited to characteristics present in computational models, so agents mental models and their corresponding social behaviors are naturally reality sim plifications.
- 1 April 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 66-73
Bistable memory and binary counters in spiking neural network
Information can be encoded in spiking neural network (SNN) by precise spike-time relations. This hypothesis can explain cell assembly formation, such as polychronous group (PNG), a notion created to explain how groups of neurons fire time-locked to each other, not necessarily synchronously. In this paper we present a set of PNGs capable of retaining triggering events in bistable states. Triggering events may be data or computational controls. Both, data and control signals are memorized as a result of intrinsic operational PNG attributes, and no neural plasticity mechanisms are involved. This behavior can be fundamental for several computational operations in SNNs. It is shown how bistable neural pools can perform tasks such as binary and stack-like counting, and how they can realize hierarchical organization in parallel computing.
- 1 April 2011
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 66-73
Bistable memory and binary counters in spiking neural network
Information can be encoded in spiking neural network (SNN) by precise spike-time relations. This hypothesis can explain cell assembly formation, such as polychronous group (PNG), a notion created to explain how groups of neurons fire time-locked to each other, not necessarily synchronously. In this paper we present a set of PNGs capable of retaining triggering events in bistable states. Triggering events may be data or computational controls. Both, data and control signals are memorized as a result of intrinsic operational PNG attributes, and no neural plasticity mechanisms are involved. This behavior can be fundamental for several computational operations in SNNs. It is shown how bistable neural pools can perform tasks such as binary and stack-like counting, and how they can realize hierarchical organization in parallel computing.
- 1 January 2011
- Published by Springer Nature
- p. 582-593
Spatial Representation and Reasoning: Integrating Synthetic Character’s Visual Perception and Symbolic Representation
- 1 January 2011
- Published by Springer Nature
- p. 582-593
Spatial Representation and Reasoning: Integrating Synthetic Character’s Visual Perception and Symbolic Representation
- Open Access
- 1 January 2011
- Published by Wiley in Mathematical Problems in Engineering
- Vol. 2011 (1)
Combining Legendre′s Polynomials and Genetic Algorithm in the Solution of Nonlinear Initial‐Value Problems
This work develops a method for solving ordinary differential equations, that is, initial‐value problems, with solutions approximated by using Legendre′s polynomials. An iterative procedure for the adjustment of the polynomial coefficients is developed, based on the genetic algorithm. This procedure is applied to several examples providing comparisons between its results and the best polynomial fitting when numerical solutions by the traditional Runge‐Kutta or Adams methods are available. The resulting algorithm provides reliable solutions even if the numerical solutions are not available, that is, when the mass matrix is singular or the equation produces unstable running processes.
- 1 January 2011
- Published by Wiley in Mathematical Problems in Engineering
- Vol. 2011 (1)
Open AccessCombining Legendre′s Polynomials and Genetic Algorithm in the Solution of Nonlinear Initial‐Value Problems
This work develops a method for solving ordinary differential equations, that is, initial‐value problems, with solutions approximated by using Legendre′s polynomials. An iterative procedure for the adjustment of the polynomial coefficients is developed, based on the genetic algorithm. This procedure is applied to several examples providing comparisons between its results and the best polynomial fitting when numerical solutions by the traditional Runge‐Kutta or Adams methods are available. The resulting algorithm provides reliable solutions even if the numerical solutions are not available, that is, when the mass matrix is singular or the equation produces unstable running processes.