Reduced deployment and calibration requirements are key for scalable and cost-effective indoor po... more Reduced deployment and calibration requirements are key for scalable and cost-effective indoor positioning systems. In this work, we propose a low-complexity, weak calibration procedure for an indoor positioning system based on infrastructure lighting and a positioning-sensitive detector. The proposed calibration relies on genetic algorithms to obtain the relevant system parameters in the real positioning environment without a priori information, and requires a low number of simple measurements. The achievable performance of the proposal was assessed by direct comparison with a formal offline calibration method requiring complex dedicated infrastructure and instruments. The comparative error assessment showed that the maximum accuracy reduction compared to the significantly more costly formal calibration was below 25 mm, and the overall absolute positioning error was smaller than 35 mm with orientation errors of around 0.25°. The performance achieved with the proposed weak calibrati...
The number of patients living with diabetes has increased significantly in recent years due to se... more The number of patients living with diabetes has increased significantly in recent years due to several factors. Many of these patients are choosing to use insulin pumps for their treatment, artificial systems that administer their insulin and consist of a glucometer and an automatic insulin supply working in an open loop. Currently, only a few closed-loop insulin delivery devices are commercially available. The most widespread systems among patients are what have been called the “Do-It-Yourself Hybrid Closed-Loop systems.” These systems require the use of platforms with high computing power. In this paper, we will present a novel wearable system for insulin delivery that reduces the energy and computing consumption of the platform without affecting the computation requirements. Patients’ information is obtained from a commercial continuous glucose sensor and a commercial insulin pump operating in a conventional manner. An ad-hoc embedded system will connect with the pump and the sen...
Here, we present an indoor positioning system (IPS) for detecting mobile agents based on a single... more Here, we present an indoor positioning system (IPS) for detecting mobile agents based on a single Position Sensitive Device sensor (PSD) sited in the environment and InfraRed Emitter Diode (IRED) located on mobile agents. The main goal of the work is to develop an alternative IPS to other sensing technologies, cheaper, easier to install and with a low computational load to obtain a high rate of measurements per second. The proposed IPS has the capacity to accurately determine 3D position from the angle of arrival (AoA) of the signal received at the PSD sensor. In this first approach to the method, the agents are considered to move along a plane. We propose two alternatives for determining position: in one, tones are emitted on a frequency unique to each transmitter, while in the other, sequences are emitted.The paper proposes and set up a very simple and easy to deploy system capable of performing 3D positioning with a single analog sensor, obtaining a high accurate positioning and a reduced execution time for the signal processing. The low computational load of the IPS makes it possible to obtain a very high position update rate (more than 100 times per second), yielding millimetric accuracies.
Here, we propose a mathematical model and a calibration procedure for a PSD (position sensitive d... more Here, we propose a mathematical model and a calibration procedure for a PSD (position sensitive device) sensor equipped with an optical system, to enable accurate measurement of the angle of arrival of one or more beams of light emitted by infrared (IR) transmitters located at distances of between 4 and 6 m. To achieve this objective, it was necessary to characterize the intrinsic parameters that model the system and obtain their values. This first approach was based on a pin-hole model, to which system nonlinearities were added, and this was used to model the points obtained with the nA currents provided by the PSD. In addition, we analyzed the main sources of error, including PSD sensor signal noise, gain factor imbalances and PSD sensor distortion. The results indicated that the proposed model and method provided satisfactory calibration and yielded precise parameter values, enabling accurate measurement of the angle of arrival with a low degree of error, as evidenced by the experimental results.
This paper focuses on optimal sensor deployment for indoor localization with a multi-objective ev... more This paper focuses on optimal sensor deployment for indoor localization with a multi-objective evolutionary algorithm. Our goal is to obtain an algorithm to deploy sensors taking the number of sensors, accuracy and coverage into account. Contrary to most works in the literature, we consider the presence of obstacles in the region of interest (ROI) that can cause occlusions between the target and some sensors. In addition, we aim to obtain all of the Pareto optimal solutions regarding the number of sensors, coverage and accuracy. To deal with a variable number of sensors, we add speciation and structural mutations to the well-known non-dominated sorting genetic algorithm (NSGA-II). Speciation allows one to keep the evolution of sensor sets under control and to apply genetic operators to them so that they compete with other sets of the same size. We show some case studies of the sensor placement of an infrared range-difference indoor positioning system with a fairly complex model of t...
Journal of Visual Communication and Image Representation, 2016
Most of the open challenges in person re-identification arise from the large variations of human ... more Most of the open challenges in person re-identification arise from the large variations of human poses as captured by different camera views. To tackle the re-identification challenges state-of-the-art methods propose to directly match image features or to learn the transformation of features that undergoes between two cameras. Other methods attack the problem by learning optimal similarity measures. However, the performance of all these methods are strongly dependent from the person pose and orientation. We focus on this aspect and introduce three main contributions ot the field: (i) propose a method to extract multiple frames of the same person with different orientation so as to capture the complete person appearance; (ii) learn the pairwise feature dissimilarities space (PFDS) formed by the subspace of pairwise feature dissimilarities computed between images of persons with similar orientation and the subspace of pairwise feature dissimilarities computed between images of persons with different orientations; (iii) within each subspace, a classifier is trained to capture the multi-modal inter-camera transformation of pairwise image dissimilarities and to discriminate between images of the same person (positive pair) and images of different persons (negative pair). To validate the proposed approach we show the superior performance of our approach to state-of-the-art methods using two publicly available benchmark datasets.
Se presenta una tarjeta electrónica diseñada para la realización de prácticas de telemetría y con... more Se presenta una tarjeta electrónica diseñada para la realización de prácticas de telemetría y control de sistemas, utilizando diferentes tecnologías de comunicaciones. El núcleo de la tarjeta lo constituye un microcontrolador ATMega 128 y en la misma se han incorporado diversas posibilidades de comunicación: Bluetooth, RS232, USB, E/S analógicas y digitales. A nivel docente su uso puede ser útil en prácticas de diseño y programación de sistemas digitales, implementación de una red de telemedicina, supervisión de sistemas (automóviles, máquinas de vending, etc), sistemas de comunicaciones etc. A nivel práctico se dispone de un circuito realizado con tecnología SMD, siendo posible su utilización en prácticas reales, constituyendo una herramienta didáctica de aplicación inmediata.
The subspace-based methods are effectively applied to classify sets of feature vectors by modelli... more The subspace-based methods are effectively applied to classify sets of feature vectors by modelling them as subspaces. However, their application to the field of non-cooperative target identification of flying aircraft is barely seen in the literature. In these methods, setting the subspace dimensionality is always an issue. Here, it is demonstrated that a modified mutual subspace method, which uses softweights to set the importance of each subspace basis, is a promising classifier for identifying sets of range profiles coming from real in-flight targets with no need to set the subspace dimensionality in advance. The assembly of a recognition database is also a challenging task. In this study, this database comprises predicted range profiles coming from electromagnetic simulations. Even though the predicted and actual profiles differ, the high recognition rates achieved reveal that the algorithm might be a good candidate for its application in an operational target recognition system.
2007 IEEE International Symposium on Industrial Electronics, 2007
This paper describes a highly efficient and parallelized HW implementation of multiple modules pr... more This paper describes a highly efficient and parallelized HW implementation of multiple modules processing a pattern matching algorithm of a 2D affine model object applied to the detection and tracking of an autonomous vehicle. The pattern matching problem has been adapted to the edge detection of the rear side of a Cycab vehicle modeled by the centroid and scale parameters. Reordering of the pattern and online generation of pixels of interest are the main contributions. The computing time for an image is independent of the number of pattern matching processing modules instantiated in hardware, making feasible to use them in large filters. The maximum operation frequency achieved has reached up to 103.1 MHz with minimal use of resources. An implementation of 200 parallel modules is capable to process 200 frames per second for image sizes of 512x512 pixels while equivalent program in a desktop PC executes only 4 fps.
Different solutions have been proposed for multiple objects tracking based on probabilistic algor... more Different solutions have been proposed for multiple objects tracking based on probabilistic algorithms. In this chapter, the authors propose the use of a single particle filter to track a variable number of objects in a complex environment. Estimator robustness and adaptability are both increased by the use of a clustering algorithm. Measurements used in the tracking process are extracted from a stereovision system, and thus, the 3D position of the tracked objects is obtained at each time step. As a proof of concept, real results are obtained in a long sequence with a mobile robot moving in a cluttered scene.
Proceedings of International Electronic Conference on Sensors and Applications, 2014
Target recognition based on radar signatures such as high resolution range profiles is a key rese... more Target recognition based on radar signatures such as high resolution range profiles is a key research domain in the Defense industry. This paper presents an automatic target recognition method based on the application of Singular Value Decomposition to a matrix of range profiles in order to reduce dimensionality, extract the main features and apply recognition in the transformed domain. So as to confirm the feasibility of the methodology, identification experiments of real in-flight targets against a database of simulated aircraft are conducted. Results show good average recognition rates implying SVD is a proper tool for recognition experiments.
This paper presents a methodology for high resolution radar image generation and automatic target... more This paper presents a methodology for high resolution radar image generation and automatic target recognition emphasizing the computational cost involved in the process. In order to obtain focused inverse synthetic aperture radar (ISAR) images certain signal processing algorithms must be applied to the information sensed by the radar. From actual data collected by radar the stages and algorithms needed to obtain ISAR images are revised, including high resolution range profile generation, motion compensation and ISAR formation. Target recognition is achieved by comparing the generated set of actual ISAR images with a database of ISAR images generated by electromagnetic software. High resolution radar image generation and target recognition processes are burdensome and time consuming, so to determine the most suitable implementation platform the analysis of the computational complexity is of great interest. To this end and since target identification must be completed in real time, computational burden of both processes the generation and comparison with a database is explained separately. Conclusions are drawn about implementation platforms and calculation efficiency in order to reduce time consumption in a possible future implementation.
The emergence of intelligent transport systems has brought out a new set of requirements on wirel... more The emergence of intelligent transport systems has brought out a new set of requirements on wireless communication. To cope with these requirements, several proposals are currently under discussion. In this highly mobile environment, the design of a prompt, efficient, flexible, and reliable medium access control, able to cover the specific constraints of the named real-time communications applications, is still unsolved. This paper presents the original proposal integrating Non-Cooperative Cognitive Time Division Multiple Access (NCC-TDMA) based on Cognitive Radio (CR) techniques to obtain a mechanism which complies with the requirements of real-time communications. Though the proposed MAC uses a slotted channel, it can be adapted to operate on the physical layer of different standards. The authors’ analysis considers the IEEE WAVE and 802.11p as the standards of reference. The mechanism also offers other advantages, such as avoiding signalling and the adaptation capacity to channel...
RESUMEN El presente trabajo tiene como objeto, mostrar diversas alternativas implementadas en lab... more RESUMEN El presente trabajo tiene como objeto, mostrar diversas alternativas implementadas en laboratorios de diseño electrónico con dispositivos FPGA's (Field Programmable Gate Array), de las titulaciones de Ingeniería Técnica de ...
2006 International Conference on Field Programmable Logic and Applications, 2006
This work shows a modular architecture based on FPGA's to solve the eigenvalue problem according ... more This work shows a modular architecture based on FPGA's to solve the eigenvalue problem according to the Jacobi method. This method is able to solve the eigenvalues and eigenvectors concurrently. The main contribution of this work is the low execution time compared with other sequential algorithms, and minimal internal FPGA consumed resources, mainly due to the fact of using the CORDIC algorithm. Two CORDIC modules have been designed to solve the trigonometric operations involved. A parallel CORDIC architecture is proposed as it is the best option to compute the eigenvalues with this method. Both CORDIC modules can work in rotation and vector mode. The whole system has been done in VHDL language, attempting to optimize the design. * Project SILPAR (Ministerio de Ciencia y Tecnología ref: DPI2003-05067) and "Cátedra de control electrónico en transportes" founded by LOGYTEL and RENFE.
The deployment of wireless sensor networks (WSNs) is a realistic solution for many markets, such ... more The deployment of wireless sensor networks (WSNs) is a realistic solution for many markets, such as manufacturing and environment monitoring, military and critical infrastructure monitoring, and, more recently, in energy-efficiency and healthcare sectors, due to their great capabilities in acquiring and transmitting data and processing them for different purposes. Current designs and architectures use radio channel(s) to share information between nodes and a gateway/hub, and implement embedded sensors with autonomous battery or low power microprocessors. Standard platforms, such as Telos B, and operating systems, such as TinyOS or Contiki, are used by a majority of stakeholders. Security, network topology, and communication protocol are critical issues in the current deployment of WSN applications. Different strategies should be developed according to the application requirements, such as distance, number of transmissions during a period of time, authentication needs, and rate of the frequency band, to name a few. This Special Issue is aimed at fostering the latest developments in the design, implementation, and evaluation in the field of WSN deployments. Summary of the Special Issue This Special Issue has focused attention on the research lines related to wireless sensor network (WSN) deployment, such as new contributions to better sensor nodes placement, enhanced routing algorithms, and alternatives to obtaining increased security. Sensor placement is an important task not fully studied in WSN applications. The number of sensors and their location will affect the performance, accuracy, and cost of the deployment. Thus, different relevant papers have been published in this particular topic, specifically WSN setups for indoor positioning. Routing is another one of the most noticeable research challenges in WSNs. Large WSNs depend on routing protocol to perform correctly. This Special Issue focuses on papers describing how the path is built by selecting appropriate nodes that are traversed along the network while maintaining a safe/secure communication. Finally, security and authentication in WSNs represent another challenging topic. Nodes in a WSNs can dynamically enter or leave a network, which leads to a variable network topology. Moreover, there are legitimate and eavesdropper nodes operating at the same time in a WSN. The published papers present contributions to balance the different tradeoffs for security and WSN shortcuts providing closed-form equations to quantify and compare the different scenarios, allowing the designer to choose the best alternative. Relevant Contributions Related to Sensor Placement Sensor placement is an important task in sensor network-based applications such as the design of indoor positioning systems, since the number of sensors and their location affect the accuracy and cost
We present a hardware implementation of the Jacobi algorithm to compute the eigenvalue decomposit... more We present a hardware implementation of the Jacobi algorithm to compute the eigenvalue decomposition (EVD). The computation of eigenvalues and eigenvectors has many applications where real time processing is required, and thus hardware implementations are often mandatory. Some of these implementations have been carried out with field programmable gate array (FPGA) devices using low level register transfer level (RTL) languages. In the present study, we used the Xilinx Vivado HLS tool to develop a high level synthesis (HLS) design and evaluated different hardware architectures. After analyzing the design for different input matrix sizes and various hardware configurations, we compared it with the results of other studies reported in the literature, concluding that although resource usage may be higher when HLS tools are used, the design performance is equal to or better than low level hardware designs.
Electric Vehicle Consumption Estimation based on Heuristics and MLP Artificial Neural Network
ABSTRACT Driving range is one of the biggest weaknesses of current electric vehicles (EV) and it ... more ABSTRACT Driving range is one of the biggest weaknesses of current electric vehicles (EV) and it is a critical parameter for planning trips. Modelling of vehicle consumption for a given road route can be achieved by analytical physical equations, whose parameters need to be adjusted for the characteristics of the vehicle and the road, and also the driver behaviour. These parameters should be revised periodically because of the changing conditions caused by car and battery aging. This paper presents a heuristic model of battery consumption based on Artificial Neural Networks (ANN) which can adapt easily to the changing conditions of the road, car and driver. The inputs to the ANN model come from a trip planning application based on Google Maps tools. The global application obtains the directions of the route given the start and end point of the trip, predicts the driving speed profile and provides an accurate estimation of the consumption through the route. The system was tested with a big collection of data provided by the Going Green Company, which was registered by eight electric vehicles travelling along six Spanish cities.
Revista Iberoamericana de Automática e Informática Industrial RIAI, 2012
Los sistemas de conteo de personas son extensamente utilizados en aplicaciones de vigilancia. En ... more Los sistemas de conteo de personas son extensamente utilizados en aplicaciones de vigilancia. En este artículo se presenta una aplicación para realizar conteo de personas a través de un sistema de estereovisión. Este sistema obtiene tasas de conteo de las personas en movimiento que atraviesan la zona de conteo recogida por el sistema estéreo distinguiendo entrada y salida. Para realizar este conteo se precisan dos fases fundamentales: detección y seguimiento. La detección se basa en la búsqueda de las cabezas de las personas por medio de una correlación de la imagen preprocesada con distintos patrones circulares, filtrando dichas detecciones por estereovisión en función de la altura. El seguimiento se lleva a cabo mediante una algoritmo de múltiples hipótesis basado en filtro de Kalman. Porúltimo, se realiza el conteo según el camino seguido por las trayectorias. Se ha experimentado con un conjunto de vídeos reales tomados en distintas zonas de tránsito en interiores de edificios, alcanzando tasas que oscilan entre un 87 % y un 98 % de acierto según la cantidad de flujo de personas que atraviesan la zona de conteo de forma simultánea. En los distintos vídeos utilizados como prueba se han reproducido todo tipo de situaciones adversas, como oclusiones, personas en grupo en diferentes sentidos, cambios de iluminación, etc.
Reduced deployment and calibration requirements are key for scalable and cost-effective indoor po... more Reduced deployment and calibration requirements are key for scalable and cost-effective indoor positioning systems. In this work, we propose a low-complexity, weak calibration procedure for an indoor positioning system based on infrastructure lighting and a positioning-sensitive detector. The proposed calibration relies on genetic algorithms to obtain the relevant system parameters in the real positioning environment without a priori information, and requires a low number of simple measurements. The achievable performance of the proposal was assessed by direct comparison with a formal offline calibration method requiring complex dedicated infrastructure and instruments. The comparative error assessment showed that the maximum accuracy reduction compared to the significantly more costly formal calibration was below 25 mm, and the overall absolute positioning error was smaller than 35 mm with orientation errors of around 0.25°. The performance achieved with the proposed weak calibrati...
The number of patients living with diabetes has increased significantly in recent years due to se... more The number of patients living with diabetes has increased significantly in recent years due to several factors. Many of these patients are choosing to use insulin pumps for their treatment, artificial systems that administer their insulin and consist of a glucometer and an automatic insulin supply working in an open loop. Currently, only a few closed-loop insulin delivery devices are commercially available. The most widespread systems among patients are what have been called the “Do-It-Yourself Hybrid Closed-Loop systems.” These systems require the use of platforms with high computing power. In this paper, we will present a novel wearable system for insulin delivery that reduces the energy and computing consumption of the platform without affecting the computation requirements. Patients’ information is obtained from a commercial continuous glucose sensor and a commercial insulin pump operating in a conventional manner. An ad-hoc embedded system will connect with the pump and the sen...
Here, we present an indoor positioning system (IPS) for detecting mobile agents based on a single... more Here, we present an indoor positioning system (IPS) for detecting mobile agents based on a single Position Sensitive Device sensor (PSD) sited in the environment and InfraRed Emitter Diode (IRED) located on mobile agents. The main goal of the work is to develop an alternative IPS to other sensing technologies, cheaper, easier to install and with a low computational load to obtain a high rate of measurements per second. The proposed IPS has the capacity to accurately determine 3D position from the angle of arrival (AoA) of the signal received at the PSD sensor. In this first approach to the method, the agents are considered to move along a plane. We propose two alternatives for determining position: in one, tones are emitted on a frequency unique to each transmitter, while in the other, sequences are emitted.The paper proposes and set up a very simple and easy to deploy system capable of performing 3D positioning with a single analog sensor, obtaining a high accurate positioning and a reduced execution time for the signal processing. The low computational load of the IPS makes it possible to obtain a very high position update rate (more than 100 times per second), yielding millimetric accuracies.
Here, we propose a mathematical model and a calibration procedure for a PSD (position sensitive d... more Here, we propose a mathematical model and a calibration procedure for a PSD (position sensitive device) sensor equipped with an optical system, to enable accurate measurement of the angle of arrival of one or more beams of light emitted by infrared (IR) transmitters located at distances of between 4 and 6 m. To achieve this objective, it was necessary to characterize the intrinsic parameters that model the system and obtain their values. This first approach was based on a pin-hole model, to which system nonlinearities were added, and this was used to model the points obtained with the nA currents provided by the PSD. In addition, we analyzed the main sources of error, including PSD sensor signal noise, gain factor imbalances and PSD sensor distortion. The results indicated that the proposed model and method provided satisfactory calibration and yielded precise parameter values, enabling accurate measurement of the angle of arrival with a low degree of error, as evidenced by the experimental results.
This paper focuses on optimal sensor deployment for indoor localization with a multi-objective ev... more This paper focuses on optimal sensor deployment for indoor localization with a multi-objective evolutionary algorithm. Our goal is to obtain an algorithm to deploy sensors taking the number of sensors, accuracy and coverage into account. Contrary to most works in the literature, we consider the presence of obstacles in the region of interest (ROI) that can cause occlusions between the target and some sensors. In addition, we aim to obtain all of the Pareto optimal solutions regarding the number of sensors, coverage and accuracy. To deal with a variable number of sensors, we add speciation and structural mutations to the well-known non-dominated sorting genetic algorithm (NSGA-II). Speciation allows one to keep the evolution of sensor sets under control and to apply genetic operators to them so that they compete with other sets of the same size. We show some case studies of the sensor placement of an infrared range-difference indoor positioning system with a fairly complex model of t...
Journal of Visual Communication and Image Representation, 2016
Most of the open challenges in person re-identification arise from the large variations of human ... more Most of the open challenges in person re-identification arise from the large variations of human poses as captured by different camera views. To tackle the re-identification challenges state-of-the-art methods propose to directly match image features or to learn the transformation of features that undergoes between two cameras. Other methods attack the problem by learning optimal similarity measures. However, the performance of all these methods are strongly dependent from the person pose and orientation. We focus on this aspect and introduce three main contributions ot the field: (i) propose a method to extract multiple frames of the same person with different orientation so as to capture the complete person appearance; (ii) learn the pairwise feature dissimilarities space (PFDS) formed by the subspace of pairwise feature dissimilarities computed between images of persons with similar orientation and the subspace of pairwise feature dissimilarities computed between images of persons with different orientations; (iii) within each subspace, a classifier is trained to capture the multi-modal inter-camera transformation of pairwise image dissimilarities and to discriminate between images of the same person (positive pair) and images of different persons (negative pair). To validate the proposed approach we show the superior performance of our approach to state-of-the-art methods using two publicly available benchmark datasets.
Se presenta una tarjeta electrónica diseñada para la realización de prácticas de telemetría y con... more Se presenta una tarjeta electrónica diseñada para la realización de prácticas de telemetría y control de sistemas, utilizando diferentes tecnologías de comunicaciones. El núcleo de la tarjeta lo constituye un microcontrolador ATMega 128 y en la misma se han incorporado diversas posibilidades de comunicación: Bluetooth, RS232, USB, E/S analógicas y digitales. A nivel docente su uso puede ser útil en prácticas de diseño y programación de sistemas digitales, implementación de una red de telemedicina, supervisión de sistemas (automóviles, máquinas de vending, etc), sistemas de comunicaciones etc. A nivel práctico se dispone de un circuito realizado con tecnología SMD, siendo posible su utilización en prácticas reales, constituyendo una herramienta didáctica de aplicación inmediata.
The subspace-based methods are effectively applied to classify sets of feature vectors by modelli... more The subspace-based methods are effectively applied to classify sets of feature vectors by modelling them as subspaces. However, their application to the field of non-cooperative target identification of flying aircraft is barely seen in the literature. In these methods, setting the subspace dimensionality is always an issue. Here, it is demonstrated that a modified mutual subspace method, which uses softweights to set the importance of each subspace basis, is a promising classifier for identifying sets of range profiles coming from real in-flight targets with no need to set the subspace dimensionality in advance. The assembly of a recognition database is also a challenging task. In this study, this database comprises predicted range profiles coming from electromagnetic simulations. Even though the predicted and actual profiles differ, the high recognition rates achieved reveal that the algorithm might be a good candidate for its application in an operational target recognition system.
2007 IEEE International Symposium on Industrial Electronics, 2007
This paper describes a highly efficient and parallelized HW implementation of multiple modules pr... more This paper describes a highly efficient and parallelized HW implementation of multiple modules processing a pattern matching algorithm of a 2D affine model object applied to the detection and tracking of an autonomous vehicle. The pattern matching problem has been adapted to the edge detection of the rear side of a Cycab vehicle modeled by the centroid and scale parameters. Reordering of the pattern and online generation of pixels of interest are the main contributions. The computing time for an image is independent of the number of pattern matching processing modules instantiated in hardware, making feasible to use them in large filters. The maximum operation frequency achieved has reached up to 103.1 MHz with minimal use of resources. An implementation of 200 parallel modules is capable to process 200 frames per second for image sizes of 512x512 pixels while equivalent program in a desktop PC executes only 4 fps.
Different solutions have been proposed for multiple objects tracking based on probabilistic algor... more Different solutions have been proposed for multiple objects tracking based on probabilistic algorithms. In this chapter, the authors propose the use of a single particle filter to track a variable number of objects in a complex environment. Estimator robustness and adaptability are both increased by the use of a clustering algorithm. Measurements used in the tracking process are extracted from a stereovision system, and thus, the 3D position of the tracked objects is obtained at each time step. As a proof of concept, real results are obtained in a long sequence with a mobile robot moving in a cluttered scene.
Proceedings of International Electronic Conference on Sensors and Applications, 2014
Target recognition based on radar signatures such as high resolution range profiles is a key rese... more Target recognition based on radar signatures such as high resolution range profiles is a key research domain in the Defense industry. This paper presents an automatic target recognition method based on the application of Singular Value Decomposition to a matrix of range profiles in order to reduce dimensionality, extract the main features and apply recognition in the transformed domain. So as to confirm the feasibility of the methodology, identification experiments of real in-flight targets against a database of simulated aircraft are conducted. Results show good average recognition rates implying SVD is a proper tool for recognition experiments.
This paper presents a methodology for high resolution radar image generation and automatic target... more This paper presents a methodology for high resolution radar image generation and automatic target recognition emphasizing the computational cost involved in the process. In order to obtain focused inverse synthetic aperture radar (ISAR) images certain signal processing algorithms must be applied to the information sensed by the radar. From actual data collected by radar the stages and algorithms needed to obtain ISAR images are revised, including high resolution range profile generation, motion compensation and ISAR formation. Target recognition is achieved by comparing the generated set of actual ISAR images with a database of ISAR images generated by electromagnetic software. High resolution radar image generation and target recognition processes are burdensome and time consuming, so to determine the most suitable implementation platform the analysis of the computational complexity is of great interest. To this end and since target identification must be completed in real time, computational burden of both processes the generation and comparison with a database is explained separately. Conclusions are drawn about implementation platforms and calculation efficiency in order to reduce time consumption in a possible future implementation.
The emergence of intelligent transport systems has brought out a new set of requirements on wirel... more The emergence of intelligent transport systems has brought out a new set of requirements on wireless communication. To cope with these requirements, several proposals are currently under discussion. In this highly mobile environment, the design of a prompt, efficient, flexible, and reliable medium access control, able to cover the specific constraints of the named real-time communications applications, is still unsolved. This paper presents the original proposal integrating Non-Cooperative Cognitive Time Division Multiple Access (NCC-TDMA) based on Cognitive Radio (CR) techniques to obtain a mechanism which complies with the requirements of real-time communications. Though the proposed MAC uses a slotted channel, it can be adapted to operate on the physical layer of different standards. The authors’ analysis considers the IEEE WAVE and 802.11p as the standards of reference. The mechanism also offers other advantages, such as avoiding signalling and the adaptation capacity to channel...
RESUMEN El presente trabajo tiene como objeto, mostrar diversas alternativas implementadas en lab... more RESUMEN El presente trabajo tiene como objeto, mostrar diversas alternativas implementadas en laboratorios de diseño electrónico con dispositivos FPGA's (Field Programmable Gate Array), de las titulaciones de Ingeniería Técnica de ...
2006 International Conference on Field Programmable Logic and Applications, 2006
This work shows a modular architecture based on FPGA's to solve the eigenvalue problem according ... more This work shows a modular architecture based on FPGA's to solve the eigenvalue problem according to the Jacobi method. This method is able to solve the eigenvalues and eigenvectors concurrently. The main contribution of this work is the low execution time compared with other sequential algorithms, and minimal internal FPGA consumed resources, mainly due to the fact of using the CORDIC algorithm. Two CORDIC modules have been designed to solve the trigonometric operations involved. A parallel CORDIC architecture is proposed as it is the best option to compute the eigenvalues with this method. Both CORDIC modules can work in rotation and vector mode. The whole system has been done in VHDL language, attempting to optimize the design. * Project SILPAR (Ministerio de Ciencia y Tecnología ref: DPI2003-05067) and "Cátedra de control electrónico en transportes" founded by LOGYTEL and RENFE.
The deployment of wireless sensor networks (WSNs) is a realistic solution for many markets, such ... more The deployment of wireless sensor networks (WSNs) is a realistic solution for many markets, such as manufacturing and environment monitoring, military and critical infrastructure monitoring, and, more recently, in energy-efficiency and healthcare sectors, due to their great capabilities in acquiring and transmitting data and processing them for different purposes. Current designs and architectures use radio channel(s) to share information between nodes and a gateway/hub, and implement embedded sensors with autonomous battery or low power microprocessors. Standard platforms, such as Telos B, and operating systems, such as TinyOS or Contiki, are used by a majority of stakeholders. Security, network topology, and communication protocol are critical issues in the current deployment of WSN applications. Different strategies should be developed according to the application requirements, such as distance, number of transmissions during a period of time, authentication needs, and rate of the frequency band, to name a few. This Special Issue is aimed at fostering the latest developments in the design, implementation, and evaluation in the field of WSN deployments. Summary of the Special Issue This Special Issue has focused attention on the research lines related to wireless sensor network (WSN) deployment, such as new contributions to better sensor nodes placement, enhanced routing algorithms, and alternatives to obtaining increased security. Sensor placement is an important task not fully studied in WSN applications. The number of sensors and their location will affect the performance, accuracy, and cost of the deployment. Thus, different relevant papers have been published in this particular topic, specifically WSN setups for indoor positioning. Routing is another one of the most noticeable research challenges in WSNs. Large WSNs depend on routing protocol to perform correctly. This Special Issue focuses on papers describing how the path is built by selecting appropriate nodes that are traversed along the network while maintaining a safe/secure communication. Finally, security and authentication in WSNs represent another challenging topic. Nodes in a WSNs can dynamically enter or leave a network, which leads to a variable network topology. Moreover, there are legitimate and eavesdropper nodes operating at the same time in a WSN. The published papers present contributions to balance the different tradeoffs for security and WSN shortcuts providing closed-form equations to quantify and compare the different scenarios, allowing the designer to choose the best alternative. Relevant Contributions Related to Sensor Placement Sensor placement is an important task in sensor network-based applications such as the design of indoor positioning systems, since the number of sensors and their location affect the accuracy and cost
We present a hardware implementation of the Jacobi algorithm to compute the eigenvalue decomposit... more We present a hardware implementation of the Jacobi algorithm to compute the eigenvalue decomposition (EVD). The computation of eigenvalues and eigenvectors has many applications where real time processing is required, and thus hardware implementations are often mandatory. Some of these implementations have been carried out with field programmable gate array (FPGA) devices using low level register transfer level (RTL) languages. In the present study, we used the Xilinx Vivado HLS tool to develop a high level synthesis (HLS) design and evaluated different hardware architectures. After analyzing the design for different input matrix sizes and various hardware configurations, we compared it with the results of other studies reported in the literature, concluding that although resource usage may be higher when HLS tools are used, the design performance is equal to or better than low level hardware designs.
Electric Vehicle Consumption Estimation based on Heuristics and MLP Artificial Neural Network
ABSTRACT Driving range is one of the biggest weaknesses of current electric vehicles (EV) and it ... more ABSTRACT Driving range is one of the biggest weaknesses of current electric vehicles (EV) and it is a critical parameter for planning trips. Modelling of vehicle consumption for a given road route can be achieved by analytical physical equations, whose parameters need to be adjusted for the characteristics of the vehicle and the road, and also the driver behaviour. These parameters should be revised periodically because of the changing conditions caused by car and battery aging. This paper presents a heuristic model of battery consumption based on Artificial Neural Networks (ANN) which can adapt easily to the changing conditions of the road, car and driver. The inputs to the ANN model come from a trip planning application based on Google Maps tools. The global application obtains the directions of the route given the start and end point of the trip, predicts the driving speed profile and provides an accurate estimation of the consumption through the route. The system was tested with a big collection of data provided by the Going Green Company, which was registered by eight electric vehicles travelling along six Spanish cities.
Revista Iberoamericana de Automática e Informática Industrial RIAI, 2012
Los sistemas de conteo de personas son extensamente utilizados en aplicaciones de vigilancia. En ... more Los sistemas de conteo de personas son extensamente utilizados en aplicaciones de vigilancia. En este artículo se presenta una aplicación para realizar conteo de personas a través de un sistema de estereovisión. Este sistema obtiene tasas de conteo de las personas en movimiento que atraviesan la zona de conteo recogida por el sistema estéreo distinguiendo entrada y salida. Para realizar este conteo se precisan dos fases fundamentales: detección y seguimiento. La detección se basa en la búsqueda de las cabezas de las personas por medio de una correlación de la imagen preprocesada con distintos patrones circulares, filtrando dichas detecciones por estereovisión en función de la altura. El seguimiento se lleva a cabo mediante una algoritmo de múltiples hipótesis basado en filtro de Kalman. Porúltimo, se realiza el conteo según el camino seguido por las trayectorias. Se ha experimentado con un conjunto de vídeos reales tomados en distintas zonas de tránsito en interiores de edificios, alcanzando tasas que oscilan entre un 87 % y un 98 % de acierto según la cantidad de flujo de personas que atraviesan la zona de conteo de forma simultánea. En los distintos vídeos utilizados como prueba se han reproducido todo tipo de situaciones adversas, como oclusiones, personas en grupo en diferentes sentidos, cambios de iluminación, etc.
Uploads
Papers by Ignacio Bravo