Inferência de percursos dos utentes de um transporte coletivo

Trabalho Final de Mestrado para obtenção do grau de Mestre em Engenharia de Redes de Comunicações e Multimédia

Localization system for pedestrians based on sensor and information fusion

Nowadays there is an increase of location-aware mobile applications. However, these applications only retrieve location with a mobile device's GPS chip. This means that in indoor or in more dense environments these applications don't work properly. To provide location information everywhere a pedestrian Inertial Navigation System (INS) is typically used, but these systems can have a large estimation error since, in order to turn the system wearable, they use low-cost and low-power sensors. In this work a pedestrian INS is proposed, where force sensors were included to combine with the accelerometer data in order to have a better detection of the stance phase of the human gait cycle, which leads to improvements in location estimation. Besides sensor fusion an information fusion architecture is proposed, based on the information from GPS and several inertial units placed on the pedestrian body, that will be used to learn the pedestrian gait behavior to correct, in real-time, the inertial sensors errors, thus improving location estimation.

4Sensing - decentralized processing for participatory sensing data

Trabalho apresentado no âmbito do Mestrado em Engenharia Informática, como requisito parcial para obtenção do grau de Mestre em Engenharia Informática.

Análise e identificação do sinal de aceleração num pé

Com a massificação do uso da tecnologia no dia-a-dia, os sistemas de localização têm vindo a aumentar a sua popularidade, devido à grande diversidade de funcionalidades que proporcionam e aplicações a que se destinam. No entanto, a maior parte dos sistemas de posicionamento não funcionam adequadamente em ambientes indoor, impedindo o desenvolvimento de aplicações de localização nestes ambientes. Os acelerómetros são muito utilizados nos sistemas de localização inercial, pelas informações que fornecem acerca das acelerações sofridas por um corpo. Para tal, neste trabalho, recorrendo à análise do sinal de aceleração provindo de um acelerómetro, propõe-se uma técnica baseada na deteção de passos para que, em aplicações futuras, possa constituir-se como um recurso a utilizar para calcular a posição do utilizador dentro de um edifício. Neste sentido, este trabalho tem como objetivo contribuir para o desenvolvimento da análise e identificação do sinal de aceleração obtido num pé, por forma a determinar a duração de um passo e o número de passos dados. Para alcançar o objetivo de estudo foram analisados, com recurso ao Matlab, um conjunto de 12 dados de aceleração (para marcha normal, rápida e corrida) recolhidos por um sistema móvel (e provenientes de um acelerómetro). A partir deste estudo exploratório tornou-se possível apresentar um algoritmo baseado no método de deteção de pico e na utilização de filtros de mediana e Butterworth passa-baixo para a contagem de passos, que apresentou bons resultados. Por forma a validar as informações obtidas nesta fase, procedeu-se, seguidamente, à realização de um conjunto de testes experimentais a partir da recolha de 33 novos dados para a marcha e corrida. Identificaram-se o número de passos efetuados, o tempo médio de passo e da passada e a percentagem de erro como as variáveis em estudo. Obteve-se uma percentagem de erro igual a 1% para o total dos dados recolhidos de 20, 100, 500 e 1000 passos com a aplicação do método proposto para a contagem do passo. Não obstante as dificuldades observadas na análise dos sinais de aceleração relativos à corrida, o algoritmo proposto mostrou bom desempenho, conseguindo valores próximos aos esperados. Os resultados obtidos permitem afirmar que foi possível atingir-se o objetivo de estudo com sucesso. Sugere-se, no entanto, o desenvolvimento de futuras investigações de forma a alargar estes resultados em outras direções.

Step based physical activity guidelines for preschool-aged children

Objective Public health organizations recommend that preschool-aged children accumulate at least 3 h of physical activity (PA) daily. Objective monitoring using pedometers offers an opportunity to measure preschooler's PA and assess compliance with this recommendation. The purpose of this study was to derive step-based recommendations consistent with the 3 h PA recommendation for preschool-aged children. Method The study sample comprised 916 preschool-aged children, aged 3 to 6 years (mean age = 5.0 ± 0.8 years). Children were recruited from kindergartens located in Portugal, between 2009 and 2013. Children wore an ActiGraph GT1M accelerometer that measured PA intensity and steps per day simultaneously over a 7-day monitoring period. Receiver operating characteristic (ROC) curve analysis was used to identify the daily step count threshold associated with meeting the daily 3 hour PA recommendation. Results A significant correlation was observed between minutes of total PA and steps per day (r = 0.76, p < 0.001). The optimal step count for ≥ 3 h of total PA was 9099 steps per day (sensitivity (90%) and specificity (66%)) with area under the ROC curve = 0.86 (95% CI: 0.84 to 0.88). Conclusion Preschool-aged children who accumulate less than 9000 steps per day may be considered Insufficiently Active.

Range of motion measurements based on depth camera for clinical rehabilitation

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

A framework for developing motion-based games

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

Desenvolvimento de um sistema para análise de equilíbrio baseado em sensores inerciais

Dissertação de mestrado integrado em Engenharia Eletrónica Industrial e de Computadores

Automatic front-crawl temporal phase detection using adaptive filtering of inertial signals

This study introduces a novel approach for automatic temporal phase detection and inter-arm coordination estimation in front-crawl swimming using inertial measurement units (IMUs). We examined the validity of our method by comparison against a video-based system. Three waterproofed IMUs (composed of 3D accelerometer, 3D gyroscope) were placed on both forearms and the sacrum of the swimmer. We used two underwater video cameras in side and frontal views as our reference system. Two independent operators performed the video analysis. To test our methodology, seven well-trained swimmers performed three 300 m trials in a 50 m indoor pool. Each trial was in a different coordination mode quantified by the index of coordination. We detected different phases of the arm stroke by employing orientation estimation techniques and a new adaptive change detection algorithm on inertial signals. The difference of 0.2 +/- 3.9% between our estimation and video-based system in assessment of the index of coordination was comparable to experienced operators' difference (1.1 +/- 3.6%). The 95% limits of agreement of the difference between the two systems in estimation of the temporal phases were always less than 7.9% of the cycle duration. The inertial system offers an automatic easy-to-use system with timely feedback for the study of swimming.

Ampliació i millora d'un vehicle teledirigit

En el projecte Ampliació i millora d’un vehicle teledirigit s’ha dut a terme l’ampliació d’un prototip de vehicle ràdio controlat fent servir dues plaques Arduino Duemilanove. Una es situa en el comandament i l’altra en el vehicle i controlen el comportament dels dos dispositius. Es transmet la informació necessària entre elles a través de dos mòduls XBee que posteriorment se’ls hi incorpora. Les plaques fetes servir en el prototip inicial eren unes SARD-13192 de Freescale i el primer que es fa en aquest sentit és una revisió del codi font utilitzat i l’adaptació a les plaques Arduino. Un acceleròmetre ADXL335 que s’incorpora a una de les plaques permet que el prototip es pugui controlar segons la posició del comandament. A més, un cop finalitzat el nou prototip és capaç de desplaçar-se endavant i endarrere, girar aturat i en moviment, i a diferents velocitats que es representen en tot moment en uns LEDs. També guarda l’últim circuit efectuat que es pot reproduir a voluntat de l’usuari, i emmagatzema les dades del recorregut de tota una sessió per exportar a l’ordinador. Per últim s’han dut a terme les proves necessàries per constatar que totes les millores s’han implementat amb èxit.

Walking activity measured by accelerometry during respiratory rehabilitation.

PURPOSE: Activity monitoring is considered a highly relevant outcome measure of respiratory rehabilitation. This study aimed to assess the usefulness of a new accelerometric method for characterization of walking activity during a 3-week inpatient rehabilitation program. METHODS: After individual calibration of the accelerometer at different walking speeds, whole-day physical activity was recorded for 15 patients with chronic obstructive pulmonary disease on the first and the last days of the program, and for 10 healthy subjects. Data were expressed as percentage of time spent in inactivity, low level activity, and medium level activity, with the latter corresponding to usual walking speed. RESULTS: The patients spent more time being inactive and less time walking than healthy subjects. At the end of the rehabilitation program, medium level activity had increased from 4% to 7% of total recording time. However, the change was not significant after periods of imposed exercise training were excluded. Walking activity increased to a greater degree among the patients with preserved limb muscle strength at entry to the program. Although health status scores improved, the changes did not correlate with the changes in walking activity. CONCLUSION: The findings lead to the conclusion that this new accelerometric method provides detailed analysis of walking activity during respiratory rehabilitation and may represent an additional useful measure of outcome.

Evaluation of the 3D symmetry of the knees when walking using na ambulatory inertia system: results 14 months after ACL Plasty.

Purpose of the study: Reconstruction of the anterior cruciate ligament (ACL) controls laxity but does not enable restoration of strictly normal 3D kinematics. The purpose of this study was to compare the kinematics of the pathological knee with that of the healthy knee after ACL plasty. This study applied a new ambulatory system using miniature captors. Material and method: Five patients with an isolated injury of the ACL participated in this study. The patients were assessed after injury (T1), at five months (T2), and at 14 months (T3) after surgery. The assessment included laxity (KT-1000), the IKDC score and the Lysholm score. The 3D angles of the knees were measured when walking 30 m on flat ground using a system composed of to small inertia units (3D accelerometer and 3D gyroscope) and a portable recorder. Functional settings were optimised and validating to ensure easy precise measurement of the 3D angles. Symmetry of the two knees was quantified using a symmetry index (SI) (difference in amplitude normalised in relation to mean amplitude) and the correlation coefficient CC. Results: Clinical indicators improved during the follow-up (IKDC T1: 3C, 2C; T2: 5B; T3: 2A, 3B; subjective IKD: 53-95; Lysholm 67-96). Mean laxity improved from 8.6m to 2.5 mm. The gait analysis showed increased symmetry in terms of amplitude for flexion-extension (SI: −17% at T1, −1% at T2, 1% at T3), and an increase in symmetry in terms of the rotation signature (CC: 0.16 at T1, 0.99 at T2, 0.99 at T3). There was no trend to varus-valgus. Discussion: This study demonstrates the clinical application of the new ambulatory system for measuring 3D angles of the knee joint. Joint symmetry increased after ACL plasty but still showed some perturbation at 14 months. The results observed here are in agreement with the literature. Other patients and other types of gait are being analysed. Conclusion: This portable system allows gait analysis outside the laboratory, before and after ACL injury. It is very useful for follow-up after surgery.

Reduced physical activity level and cardiorespiratory fitness in children with chronic diseases.

We aimed to compare physical activity level and cardiorespiratory fitness in children with different chronic diseases, such as type 1 diabetes mellitus (T1DM), obesity (OB) and juvenile idiopathic arthritis (JIA), with healthy controls (HC). We performed a cross-sectional study including 209 children: OB: n = 45, T1DM: n = 48, JIA: n = 31, and HC: n = 85. Physical activity level was assessed by accelerometer and cardiorespiratory fitness by a treadmill test. ANOVA, linear regressions and Pearson correlations were used. Children with chronic diseases had reduced total daily physical activity counts (T1DM 497 +/- 54 cpm, p = 0.003; JIA 518 +/- 28, p < 0.001, OB 590 +/- 25, p = 0.003) and cardiorespiratory fitness (JIA 39.3 +/- 1.7, p = 0.001, OB 41.7 +/- 1.2, p = 0.020) compared to HC (668 +/- 35 cpm; 45.3 +/- 0.9 ml kg(-1) min(-1), respectively). Only 60.4% of HC, 51.6% of OB, 38.1% of JIA and 38.5% of T1DM children met the recommended daily 60 min of moderate-to-vigorous physical activity. Low cardiorespiratory fitness was associated with female gender and low daily PA. Children with chronic diseases had reduced physical activity and cardiorespiratory fitness. As the benefits of PA on health have been well demonstrated during growth, it should be encouraged in those children to prevent a reduction of cardiorespiratory fitness and the development of comorbidities.

Can accelerometry accurately predict the energy cost of uphill/downhill walking?

To evaluate whether an activity monitor based on body acceleration measurement can accurately assess the energy cost of the human locomotion, 12 subjects walked a combination of three different speeds (preferred speed +/- 1 km/h) and seven slopes (-15 to +15% by steps of 5%) on a treadmill. Body accelerations were recorded using a triaxial accelerometer attached to the low back. The mean of the integral of the vector magnitude (norm) of the accelerations (mIAN) was calculated. VO2 was measured using continuous indirect calorimetry. When the results were separately analysed for each incline, mIAN was correlated to VO2 (average r = 0.87, p<0.001, n = 36). VO2 was not significantly correlated to mIAN when data were globally analysed (n = 252). Large relative errors occurred when predicted VO2 (estimated from data of level walking) was compared with measured VO2 for different inclines (-53% at +15% incline, to +55% at -15% incline). It is concluded that without an external measurement of the slope, the standard method of analysis of body accelerations cannot accurately predict the energy cost of uphill or downhill walking.

Contribution of physical education to overall physical activity.

For many children, physical activity (PA) during physical education (PE) lessons provides an important opportunity for being physically active. Although PA during PE has been shown to be low, little is known about the contribution of PA during PE to overall PA. The aim was therefore to assess children's PA during PE and to determine the contribution of PE to overall PA with special focus on overweight children. Accelerometer measurements were done in 676 children (9.3 ± 2.1 years) over 4-7 days in 59 randomly selected classes. Moderate-and-vigorous PA (MVPA; ≥ 2000 counts/min) during PE (MVPA(PE) ), overall MVPA per day (MVPA(DAY) ), and a comparison of days with and without PE were calculated by a regression model with gender, grade, and weight status (normal vs overweight) as fixed factors and class as a random factor. Children spent 32.8 ± 15.1% of PE time in MVPA. Weight status was not associated to MVPA(PE) . MVPA(PE) accounted for 16.8 ± 8.5% of MVPA(DAY) , and 17.5 ± 8.2% in overweight children. All children were more active on days with PE than on days without PE (differences: 16.1 ± 29.0 min of MVPA(DAY) ; P ≤ 0.001; 13.7 ± 28.0 min for overweight children). Although MVPA(PE) was low, PE played a considerable role in providing PA and was not compensated by reducing extracurricular MVPA.