Wednesday 10.04.2024 at 14:00

Wolfson Building of Mechanical Engineering, Room 206

COMPLIANT PARALLEL MOTION LINKAGE AMPLIFICATION MECHANISM FOR RESONANT FORCE/ACCELRATION SENSING

Shay Goldstein

M.Sc. Student, under the supervision of Prof. Slava Krylov

School of Mechanical Engineering, Tel Aviv University, Tel Aviv, Israel

Inertial sensors such as angular rate sensors (gyroscopes) and accelerometers are among the most successful examples of applications relying on the microelectromechanical systems (MEMS) technology.   These low-cost, small size, energetically efficient low power consumption devices are widely implemented in consumer electronics, automotive, and airspace industries and serve as key components of autonomous robots and advanced high-end navigation, guidance, and control systems. Within the inertial MEMS area, resonant accelerometers attract the steadily increasing attention of researchers and MEMS developers. The operational principle of these devices is based on the monitoring of the acceleration-dependent resonant frequency of a vibrating sensing beam attached to a proof mass and stretched (or compressed) by an inertial force.  Resonant accelerometers are distinguished by a wider measurement range combined with uncompromised sensitivity, lower noise, and improved robustness when compared to the statically operated acceleration sensors. Among the challenges of resonant accelerometers’ mechanical design is the development of a force amplification mechanism usually unavoidable to reach the required sensitivity, along with mitigation of undesirable output nonlinearities and cross-axis couplings.

The present work is focused on the theoretical investigation of the vibrating beam accelerometer (VBA) implementing a recently suggested compliant parallel motion linkage amplification mechanism. The device incorporates two vibrating sensing beams attached at their ends to four proof masses. The unique compliant force amplifier architecture allows to reach high sensitivity and purely extensional, without any bending, deformation of the sensing beams, but may result in undesired nonlinearities and structural cross-couplings. The primary objective of the work was to explore the possibility of sensor performance enhancement by addressing the challenges related to the output nonlinearity and cross-axis effects mitigation. The reduced order (RO) model of the device was built using the stationary potential energy principle. The hierarchy of models included fully and weakly nonlinear and linearized models. The nonlinear equilibrium equations were solved numerically and the sensing beams’ frequency shifts due to the acceleration were calculated. The RO model predictions were compared to the results of the three-dimensional numerical finite elements analysis.  Our results show that careful mechanical design is instrumental in the achievement of the highest force amplification and therefore, the highest sensitivity, low scale factor nonlinearity, and low cross-axis sensitivity.  Our results provide useful insights extending the arsenal of tools that can be used by the designers of a large variety of dynamically operated inertial, pressure, magnetic, and electric field microsensors based on resonant frequency monitoring.

Electrical Engineering Systems Seminar

Speaker: Sapir Kontente

M.Sc. student under the supervision of Prof. Ben-Zion Bobrovsky

Wednesday, 25^th February 2024, at 15:00

Room 011, Kitot Building, Faculty of Engineering

CLRmatchNet: Enhancing Curved Lane Detection with Deep Matching Process

Abstract

Lane detection is crucial for safe autonomous navigation, providing essential data for accurate vehicle guidance. Currently, anchor-based detectors are the leading paradigm for lane detection and are commonly combined with a classical label assignment step during training, labeling model predictions for optimal learning. Accurate label assignment has great impact on the model performance, that is usually relying on a predefined classical cost function evaluating GT-prediction alignment. However, classical label assignment methods face limitations due to their reliance on predefined cost functions derived from low-dimensional models, potentially impacting their optimality.

This study introduces MatchNet, a deep learning submodule aimed at addressing challenges in the label assignment stage for lane detection. Integrated into the state-of-the-art lane detection network, CLRNet, MatchNet replaces the classic label assignment process and exhibits significant enhancements, particularly in detecting curved lanes. It raises confidence levels in positive lanes and introduces dynamic flexibility in lane matching. The integrated model, CLRmatchNet, surpasses CLRNet, showcasing substantial improvements in scenarios involving curved lanes, with enhancements of +2.8% for ResNet34, +2.3% for ResNet101, and +2.96% for DLA34.

השתתפות בסמינר תיתן קרדיט שמיעה = עפ"י רישום שם מלא + מספר ת.ז. בדף הנוכחות שיועבר באולם במהלך הסמינר

Wednesday 10.04.2024 at 14:00

Wolfson Building of Mechanical Engineering, Room 206

Assessing the Effect of Floating Photovoltaic Panels on Secondary

Effluents’ Quality – Case Study in Yesodot Reservoir, Israel

The Iby and Alder Fleischman Faculty of Engineering

Environmental Engineering program

Yosef Perlmutter

Advisor: Prof. Hadas Mamane

Israel is a leading country in wastewater treatment, where 96% of the wastewater was treated in 2020, from which more than 80% were used for irrigation. The Israeli standards for irrigation are divided into restricted and non-restricted, based on the effluents’ quality. The Israeli Ministry of Health regulated that in reservoirs with restricted irrigation effluents’ quality, only 20% of the reservoir can be covered with floating photovoltaic panels. In this study, two data loggers were submerged at 1-1.5m depth in Yesodot Reservoir, which has a floating photovoltaic panels array installed. This study’s objective is to establish an experimental setup for determining the impact of floating photovoltaic panels on a secondary effluents’ reservoir, emphasizing the end of the irrigational season. The specific objectives are to (1) determine the temporal differences in dissolved oxygen and temperature between two locations within the Yesodot Reservoir, one beneath the photovoltaic panels’ array, and the other near the reservoir’s outlet; (2) understand whether the panels cause anaerobic or low dissolved oxygen conditions and/or significant local temperature differences, and (3) conduct comparative analysis between the locations during the end of the irrigation season, while the reservoir reaches its minimal volume and surface area, correlating to the poorest effluent quality.

(The talk will be given in English)

Speaker:     Amitay Eldar

Applied Math Department, Tel Aviv University

011 hall, Electrical Engineering-Kitot Building

Monday, February 12^th, 2024

15:00 - 16:00

Abstract: 

Detecting unknown objects in noisy data is a key problem in many fields of science, such as electron microscopy imaging. A common model for the unknown objects is the linear subspace model, which assumes that the objects can be expanded in some known basis (such as the Fourier basis). In this talk, I will present an object detection algorithm that under the linear subspace model is asymptotically guaranteed to find all objects while making only a small percentage of false discoveries.

Amitay is currently a PhD student in the Applied Math department at Tel Aviv University under the supervision of Prof. Shkolnisky Yoel. He received his BSc degree in Mathematics from Tel Aviv University in 2016, and his MSc in Applied Mathematics from Tel Aviv University in 2020. His research interests are in Object Detection, Multiple Hypotheses Testing and Signal Processing.