... Fernandez, R and Zegers, P and Weber, G and Tyler, N Effect of door width, platform height an... more ... Fernandez, R and Zegers, P and Weber, G and Tyler, N Effect of door width, platform height and fare collection on bus dwell time: laboratory evidence for Santiago de Chile. Transportation Research Record (In press). Full text not available from this repository. Type: Article. ...
Alto del andén, ancho de puertas y cobro de tarifa en la demora al transporte público. Resultados... more Alto del andén, ancho de puertas y cobro de tarifa en la demora al transporte público. Resultados de experiencias en laboratorio FERNANDEZ, Rodrigo; ZEGERS, Pablo; WEBER, Gustavo; FIGUEROA, Álvaro; TYLER, Nick ... ALTO DEL ANDÉN, ANCHO DE PUERTAS Y
Designing efficient methods for training dynamic neural networks for learning spatio-temporal pat... more Designing efficient methods for training dynamic neural networks for learning spatio-temporal patterns is of great interest at present. In particular, the “trajectory generation problem” that involves training the network to learn and replicate autonomously a specified time-varying periodic motion has attracted considerable attention. A systematic approach to solve this problem by decomposing the overall task into two sub-tasks, a spatio-temporal sequence assignment and a mapping of ordered sequences, is presented. This decomposition permits the dynamic neural network to be realized as a cascade of a simple recurrent net followed by a non-recurrent one that yields considerable reduction in training complexity. A detailed performance evaluation of the present scheme is given by considering several trajectory generation experiments that highlight the strong points of this approach, which include simplicity and accuracy in training, flexibility to include control parameters in order to...
1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227)
ABSTRACT A novel system that efficiently integrates two types of neural networks for reliably per... more ABSTRACT A novel system that efficiently integrates two types of neural networks for reliably performing isolated word recognition is described. The recognition system comprises of a feature extractor that includes a self organizing map for an optimal tailoring of trajectory representations of words in reduced dimension feature spaces. Experimental results indicate that such lower dimensional trajectories can provide a reliable representation of spoken words, while reducing the training complexity for the recognition of the trajectory. A recurrent neural network is employed for performing trajectory recognition and a method that allows us to progressively grow the training set is utilized for network training. The optimal tailoring of trajectories and growing training sets are two innovations that result in a superior training of the recurrent neural network, which in turn delivers a robust word recognition performance tolerating wide variations in the speech signal
The 2006 IEEE International Joint Conference on Neural Network Proceedings, 2006
A consistent density function estimator is presented. Whether an estimator is consistent or not i... more A consistent density function estimator is presented. Whether an estimator is consistent or not is critical when the desired solution is not known. Without determining consistency it is not possible to know if the solution generated by an algorithm is close to the true solution or not. A combination of performance index, MLP architecture, training algorithm, and statistical learning theory concepts is used to produce consistent one dimensional density function estimations. The performance index and the MLP architecture are designed using information theoretical and algorithmic considerations, whereas the consistency of the solution is determined from the behavior exhibited by the estimator throughout the training process. The training algorithm is designed to highlight behavior that has been proven to exist in other learning problems with the help of statistical learning theory. The algorithm is tested with examples in order to determine the extent of its usefulness and to study its limitations.
2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541)
In training a learning machine (LM) with unlimited data samples available in the training set, it... more In training a learning machine (LM) with unlimited data samples available in the training set, it is important to be able to determine when the LM has attained an adequate level of generalization in order to stop the training process. While this is a problem that has not yet achieved a satisfactory solution, aiding the determination of the generalization level
In previous papers presented at the European Transport Conference (ETC), Fernandez (2011) shown b... more In previous papers presented at the European Transport Conference (ETC), Fernandez (2011) shown boarding and alighting times of passengers for the Transantiago system obtained at the University College London Pedestrian Accessibility and Movement Environment Laboratory (PAMELA). Previously, Fernandez et al (2010) analysed by means of a pedestrian microsimulator metro-bus interchange spaces in order to propose design guidelines, taking as case study a terminal station of Metro de Santiago. Following this line of research, the aim of this paper is the estimation of pedestrian saturation flows through public transport doors under different physical configurations by means of real-scale experiments made at PAMELA. Results presented in this paper are part of two projects funded by the Chilean Fund of Science and Technology (FONDECYT), aimed at understanding pedestrian behaviour at public transport facilities by means of artificial vision. The saturation flow is a parameter using by traffic engineers to calculate the capacity of traffic signal junctions. For a given junction approach, the saturation flow is defined as the maximum discharge rate of a queue of vehicles during the effective green time of that approach. It is well-known that at the start of the green period there is a transient period before the discharge rate reaches its maximum, which is the saturation flow for that approach. If the queue remains until the end of the green time, there is another transient period until the start of the red time. The value of the saturation flow and transient periods depends on both the traffic composition and geometric characteristics of the junction approach. The authors previous research on bus boarding and alighting times suggest that a similar behaviour occurs though public transport doors. This hypothesis was tested by re-processing the 2008 videos for different door widths and platform heights. The authors used the approach of the Road Note 34 (RRL, 1963) for measuring passenger’s saturation flow. Values of passenger saturation flows and transient periods can be used by traffic engineers to estimate delays of public transport vehicles at bus stops and metro stations. This in turn can help architects in designing pedestrian facilities at transport infrastructures. The validation of the authors hypothesis can also help transport planners with the calculation of commercial speed, fleet size, type of vehicles, and operational costs of public transport systems. The authors are conducting further real-scale experiments in the Human Dynamics Laboratory of University of Los Andes to analyse the discharge of a group of people through a public transport door with different physical configurations. we are using a mock-up of the hall of a public transport vehicle in our experiments. Some of the variables that have being studied are the effect of different door widths on passenger’s saturation flow.
The 2006 IEEE International Joint Conference on Neural Network Proceedings, 2006
ABSTRACT The finite element method frequently needs complex grids to solve partial differential e... more ABSTRACT The finite element method frequently needs complex grids to solve partial differential equations. This becomes more serious in highly dimensional problems and complicated geometries. In this article we present an improved gridless solver, which trains a neural network to fit the differential equation solution. The advantage of a gridless method is its easier scalability to problems with a high number of dimensions. A smart stopping criterion, based on statistical learning theory concepts, makes the method more autonomous than preceding algorithms. The proposed method uses a simple rule to include the boundary conditions in the error measure of the network. For validation, we show the results of solving some simple first and second order equations and one from a classical application problem.
2009 International Joint Conference on Neural Networks, 2009
A neural network training method for identification in bounded time of nonlinear systems is prese... more A neural network training method for identification in bounded time of nonlinear systems is presented in this paper. A sliding mode surface drives the adalines, perceptrons and multilayer perceptrons so as to a new second order sliding mode is enforced for all time. This neural network-based sliding mode enforces an invariant differential manifold, with a time-varying feedback gain to give
Transportation Research Record: Journal of the Transportation Research Board, 2010
Dwell time is the time that a public transport vehicle remains stopped while transferring passeng... more Dwell time is the time that a public transport vehicle remains stopped while transferring passengers. Dwell time depends on the number of boarding and alighting passengers plus other characteristics, such as platform height, door width, fare collection method, internal layout of the vehicle, and occupancy of the vehicle. Traditionally, dwell time has been described as a linear function of the number of passengers boarding and alighting. In this paper, results are presented of dwell time parameters obtained from real-scale experiments made at the Pedestrian Accessibility and Movement Environment Laboratory, University College London. Three variables were controlled: platform height (0, 150, and 300 mm), door width (800 and 1,600 mm), and fare collection method (prepayment outside the vehicle and payment with an electronic card at the entrance of the vehicle). For each value of the variables mentioned above, between 15 and 20 runs were recorded on videotape with four cameras and diffe...
... Fernandez, R and Zegers, P and Weber, G and Tyler, N Effect of door width, platform height an... more ... Fernandez, R and Zegers, P and Weber, G and Tyler, N Effect of door width, platform height and fare collection on bus dwell time: laboratory evidence for Santiago de Chile. Transportation Research Record (In press). Full text not available from this repository. Type: Article. ...
Alto del andén, ancho de puertas y cobro de tarifa en la demora al transporte público. Resultados... more Alto del andén, ancho de puertas y cobro de tarifa en la demora al transporte público. Resultados de experiencias en laboratorio FERNANDEZ, Rodrigo; ZEGERS, Pablo; WEBER, Gustavo; FIGUEROA, Álvaro; TYLER, Nick ... ALTO DEL ANDÉN, ANCHO DE PUERTAS Y
Designing efficient methods for training dynamic neural networks for learning spatio-temporal pat... more Designing efficient methods for training dynamic neural networks for learning spatio-temporal patterns is of great interest at present. In particular, the “trajectory generation problem” that involves training the network to learn and replicate autonomously a specified time-varying periodic motion has attracted considerable attention. A systematic approach to solve this problem by decomposing the overall task into two sub-tasks, a spatio-temporal sequence assignment and a mapping of ordered sequences, is presented. This decomposition permits the dynamic neural network to be realized as a cascade of a simple recurrent net followed by a non-recurrent one that yields considerable reduction in training complexity. A detailed performance evaluation of the present scheme is given by considering several trajectory generation experiments that highlight the strong points of this approach, which include simplicity and accuracy in training, flexibility to include control parameters in order to...
1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227)
ABSTRACT A novel system that efficiently integrates two types of neural networks for reliably per... more ABSTRACT A novel system that efficiently integrates two types of neural networks for reliably performing isolated word recognition is described. The recognition system comprises of a feature extractor that includes a self organizing map for an optimal tailoring of trajectory representations of words in reduced dimension feature spaces. Experimental results indicate that such lower dimensional trajectories can provide a reliable representation of spoken words, while reducing the training complexity for the recognition of the trajectory. A recurrent neural network is employed for performing trajectory recognition and a method that allows us to progressively grow the training set is utilized for network training. The optimal tailoring of trajectories and growing training sets are two innovations that result in a superior training of the recurrent neural network, which in turn delivers a robust word recognition performance tolerating wide variations in the speech signal
The 2006 IEEE International Joint Conference on Neural Network Proceedings, 2006
A consistent density function estimator is presented. Whether an estimator is consistent or not i... more A consistent density function estimator is presented. Whether an estimator is consistent or not is critical when the desired solution is not known. Without determining consistency it is not possible to know if the solution generated by an algorithm is close to the true solution or not. A combination of performance index, MLP architecture, training algorithm, and statistical learning theory concepts is used to produce consistent one dimensional density function estimations. The performance index and the MLP architecture are designed using information theoretical and algorithmic considerations, whereas the consistency of the solution is determined from the behavior exhibited by the estimator throughout the training process. The training algorithm is designed to highlight behavior that has been proven to exist in other learning problems with the help of statistical learning theory. The algorithm is tested with examples in order to determine the extent of its usefulness and to study its limitations.
2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541)
In training a learning machine (LM) with unlimited data samples available in the training set, it... more In training a learning machine (LM) with unlimited data samples available in the training set, it is important to be able to determine when the LM has attained an adequate level of generalization in order to stop the training process. While this is a problem that has not yet achieved a satisfactory solution, aiding the determination of the generalization level
In previous papers presented at the European Transport Conference (ETC), Fernandez (2011) shown b... more In previous papers presented at the European Transport Conference (ETC), Fernandez (2011) shown boarding and alighting times of passengers for the Transantiago system obtained at the University College London Pedestrian Accessibility and Movement Environment Laboratory (PAMELA). Previously, Fernandez et al (2010) analysed by means of a pedestrian microsimulator metro-bus interchange spaces in order to propose design guidelines, taking as case study a terminal station of Metro de Santiago. Following this line of research, the aim of this paper is the estimation of pedestrian saturation flows through public transport doors under different physical configurations by means of real-scale experiments made at PAMELA. Results presented in this paper are part of two projects funded by the Chilean Fund of Science and Technology (FONDECYT), aimed at understanding pedestrian behaviour at public transport facilities by means of artificial vision. The saturation flow is a parameter using by traffic engineers to calculate the capacity of traffic signal junctions. For a given junction approach, the saturation flow is defined as the maximum discharge rate of a queue of vehicles during the effective green time of that approach. It is well-known that at the start of the green period there is a transient period before the discharge rate reaches its maximum, which is the saturation flow for that approach. If the queue remains until the end of the green time, there is another transient period until the start of the red time. The value of the saturation flow and transient periods depends on both the traffic composition and geometric characteristics of the junction approach. The authors previous research on bus boarding and alighting times suggest that a similar behaviour occurs though public transport doors. This hypothesis was tested by re-processing the 2008 videos for different door widths and platform heights. The authors used the approach of the Road Note 34 (RRL, 1963) for measuring passenger’s saturation flow. Values of passenger saturation flows and transient periods can be used by traffic engineers to estimate delays of public transport vehicles at bus stops and metro stations. This in turn can help architects in designing pedestrian facilities at transport infrastructures. The validation of the authors hypothesis can also help transport planners with the calculation of commercial speed, fleet size, type of vehicles, and operational costs of public transport systems. The authors are conducting further real-scale experiments in the Human Dynamics Laboratory of University of Los Andes to analyse the discharge of a group of people through a public transport door with different physical configurations. we are using a mock-up of the hall of a public transport vehicle in our experiments. Some of the variables that have being studied are the effect of different door widths on passenger’s saturation flow.
The 2006 IEEE International Joint Conference on Neural Network Proceedings, 2006
ABSTRACT The finite element method frequently needs complex grids to solve partial differential e... more ABSTRACT The finite element method frequently needs complex grids to solve partial differential equations. This becomes more serious in highly dimensional problems and complicated geometries. In this article we present an improved gridless solver, which trains a neural network to fit the differential equation solution. The advantage of a gridless method is its easier scalability to problems with a high number of dimensions. A smart stopping criterion, based on statistical learning theory concepts, makes the method more autonomous than preceding algorithms. The proposed method uses a simple rule to include the boundary conditions in the error measure of the network. For validation, we show the results of solving some simple first and second order equations and one from a classical application problem.
2009 International Joint Conference on Neural Networks, 2009
A neural network training method for identification in bounded time of nonlinear systems is prese... more A neural network training method for identification in bounded time of nonlinear systems is presented in this paper. A sliding mode surface drives the adalines, perceptrons and multilayer perceptrons so as to a new second order sliding mode is enforced for all time. This neural network-based sliding mode enforces an invariant differential manifold, with a time-varying feedback gain to give
Transportation Research Record: Journal of the Transportation Research Board, 2010
Dwell time is the time that a public transport vehicle remains stopped while transferring passeng... more Dwell time is the time that a public transport vehicle remains stopped while transferring passengers. Dwell time depends on the number of boarding and alighting passengers plus other characteristics, such as platform height, door width, fare collection method, internal layout of the vehicle, and occupancy of the vehicle. Traditionally, dwell time has been described as a linear function of the number of passengers boarding and alighting. In this paper, results are presented of dwell time parameters obtained from real-scale experiments made at the Pedestrian Accessibility and Movement Environment Laboratory, University College London. Three variables were controlled: platform height (0, 150, and 300 mm), door width (800 and 1,600 mm), and fare collection method (prepayment outside the vehicle and payment with an electronic card at the entrance of the vehicle). For each value of the variables mentioned above, between 15 and 20 runs were recorded on videotape with four cameras and diffe...
Uploads
Papers by Pablo Zegers