Mechatronics Systems Engineering - Theses, Dissertations, and other Required Graduate Degree Essays

Receive updates for this collection

Design of Interface Circuits for Capacitive Sensing Applications

Author: 
Date created: 
2014-08-08
Abstract: 

This thesis focuses on the design of integrated readout circuits for differential capacitive sensing applications. Such circuits are needed especially for interfacing with microsensors where capacitive transduction is predominantly used. The result of this research is the development of common framework for interface circuitries suitable for different sensing applications. These interface circuits were designed and fabricated in standard Complementary Metal-Oxide-Semiconductor (CMOS) processes and can be integrated into the design of various sensing systems. The proposed circuits in this work are characterized by high dynamic range, low power consumption, and adjustable sensing range. Such circuits promote easy-to-use user interfaces while having a low cost. Three different circuit designs were proposed and form the highlights of this thesis. The first interface circuit is a novel realization of a synchronous demodulation technique. The main advantage of the proposed circuit compared to state-of-the-art is that it has a high sensing dynamic range of 112dB and is capable of measuring capacitance as small as 30 aF with a total power consumption of 8mW. Low power consumption is one of the most important design criteria for portable sensing systems besides accuracy and precision. Following this requirement, low power consumption is the main criterion in the second circuit proposed in this work. This circuit uses a switch-based capacitance-to-voltage converter that is designed and fabricated in 0.35μm CMOS technology. This circuit had a low power consumption of 600μW. Its simple structure offers area and power advantages over the more complex circuits. In addition, its ratiometric sensing feature provides an adjustable sensing range which can be tuned for different applications. This circuit can detect capacitances as small as 230 aF in 1pF range of capacitance. To reduce the effect of parasitics on the circuit performance and improve the linearity, the design of the second circuit was enhanced. By using an additional block and an analog divider, the sensitivity of the circuit to parasitics was significantly reduced. On the other hand, a time based output allowed for the elimination of the analog buffers. The fabricated circuit consumed a total power of only 720μW and was fabricated in 0.35μm CMOS technology. Another advantage of this circuit over the previous designs is that the pulse-width output signal of this circuit can be more easily digitized. The proposed circuits in this thesis have been tested with different types of sensors including humidity, motion, and variable MEMS capacitors. For all of them also, the measurement results are found to be in good agreement with the analytic and simulation results. These circuits can be used as standalone chips or can be integrated into the design of larger sensing systems.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Behraad Bahreyni
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) Ph.D.

Highly Sensitive Printed Accelerometer for Biomedical Applications

Author: 
Date created: 
2014-05-27
Abstract: 

This thesis introduces a highly sensitive single-axis printed accelerometer based on a flexible paper substrate. The accelerometer is fabricated by cost-effective silver nano ink printing technologies, and consists of a suspended parallel-plate sensing capacitor. By designing the suspension bridge and proof mass structures, the sensitivity to vertical accelerations is optimized based on simulation results. The optimized design with two long ellipse-shaped bridges exhibits a capacitive sensitivity of 20 fF/g at z-axis acceleration of 1-10 g. A wearable sensing system is proposed which composes of the light-weight flexible accelerometer integrated with a readout circuit. The bandage-type accelerometer system can be conformally attached to various positions on human body for motion detection as well as obtaining vital signs such as human pulse and respiratory rate. Thus, these features allow unobtrusive continuous monitoring of various vital signs and physical activities and acting as a multifunctional sensor in health monitoring system.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Woo Soo Kim
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Developing roll-to-roll manufacturing system for flexible and stretchable electronics by direct stamping of silver nano-ink

Author: 
Date created: 
2014-06-17
Abstract: 

Direct stamping of silver nanoparticle based ink has been developed for cost-effective and process-effective manufacturing of flexible or stretchable electronic devices. Facile removal of residual layer from deposited silver nanoparticles (AgNPs) layer results in high fidelity of final silver electrode without further post-processes. Scanning electron microscopy (SEM) images and energy dispersive x-ray spectroscopic analysis have revealed residue-free transfer of microscale inter-digitated capacitors onto flexible or stretchable substrates. Enhanced adhesion with viscous UV-curable adhesive enables perfect contact between the substrate and AgNPs in the patterned trenches. As a result, adhesion between them is maximized enough to transfer AgNPs from the trenches to the target substrates. The direct stamping also leads to densification of AgNPs in the patterned trenches by pressing force for contacting the substrate and the stamp. The densification results in lower annealing temperature (~130°C) of AgNPs than normally required annealing temperature (160°C) and better mechanical stability of resulting patterns than one from other ink-based printing techniques. SEM and atomic force microscopy show highly reduced number of pores and less roughness of the layer of AgNPs as it confirms densification of AgNPs. For further understanding of the densification in the microscopic view, simulation of the direct stamping has been performed as normal pressure is applied vertically to the stamp. The simulation reveals that by external pressure during direct stamping, the layer of AgNPs is compressed in the direction of the external pressure and is extended in the normal direction to the pressure. Extension provides room for AgNPs to move helping compression of the AgNPs’ layer by the external pressure. In addition, restoring force of the PDMS stamp enables further densification of AgNPs in the x-direction after removal of the pressure to the stamp. Finally, a table-top roll-to-roll (R2R) manufacturing system has been also successfully built to demonstrate high throughput manufacturing of flexible and stretchable electronic devices by the developed direct stamping method. Strain sensors, touch pads and radio-frequency identification (RFID) tags have been fabricated for highly sensitive flexible and stretchable sensor applications by the R2R direct stamping with the maximum speed of 1 m/minute. The fabricated sensors are still operational even with 8 mm bending radius and after 10,000 cycles of bending. And the wireless and stretchable RFID tags have shown strain sensing behavior up to 7% of stretching strain.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Woo Soo Kim
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) Ph.D.

Improving hydrogen sensors for diagnostics in heavy duty fuel cell applications

Date created: 
2014-06-02
Abstract: 

Ballard Power Systems is a leading company in the field of hydrogen fuel cell research and manufacture. Monitoring membrane deterioration through excessive crossover requires accurate hydrogen leak sensors within the vehicle safety systems. Hydrogen sensors are typically used to detect fuel crossover and external leaks in fuel cells. They are expensive and have typically exhibited short lifetimes. Two methods of lengthening the sensor life have been examined in this thesis. The first is physical filtering of poisoning agents. These components, particularly the siloxane compounds that are off-gassed from the silicone tubing, coalesce on the surface of the hydrogen sensor as silicates, reducing the sensor sensitivity. With these compounds mostly removed by the filters, the sensor life has been extended. While preventing degradation of the cathode exhaust sensor is the best approach, it was not wholly effective. Significant effort went into recalibrating the hydrogen sensor in situ, the second method. This methodology uses anode and cathode mass flow sensors in the fuel cell to automatically calibrate the hydrogen sensor as it degrades. The results from this approach were found to be promising, and we showed that the auto-calibration algorithm was robust enough to accept disturbances in system inputs.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Farid Golnaraghi
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Real-time SLAM for Humanoid Robot Navigation Using Augmented Reality

Author: 
Date created: 
2014-04-03
Abstract: 

The integration of Augmented Reality (AR) with Extended Kalman Filter based Simultaneously Localization and Mapping (EKF-SLAM) is proposed and implemented on a humanoid robot in this thesis. The goal has been to improve the performance of EKF-SLAM in terms of reducing the computational effort, to simplify the data association problem and to improve the trajectory control of the algorithm. Two applications of Augmented Reality are developed. In the first application, during a standard EKF-SLAM process, the humanoid robot recognizes specific and predefined graphical markers through its camera and obtains landmark information and navigation instruction using Augmented Reality. In the second stage, iPhone on-board gyroscope sensor is applied to achieve an enhanced positioning system, which is then used in conjunction of a PI motion controller for trajectory control. The proposed applications are implemented and verified in real-time on the humanoid robot NAO.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Ahmad Rad
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

Mechanical Characterization of Breast Tissue Constituents for Cancer Assessment

Author: 
Date created: 
2014-02-25
Abstract: 

Breast elastography is a method of cancer detection that uses the response of soft tissue to deformations, leading to discovery of abnormalities. The methods of Clinical Breast Examination and Breast Self-Examination are based primarily on stiffness and, hence, on the mechanics of tissue constituents examined by palpation (Goodson, 1996). However, little is known about the mechanical characteristics of breast tissue under compression and the contribution of tissue mechanics to breast cancer detection. This study focuses therefore on tissue characterization and on identification of the relationship between tissue properties and pathological mechanics via offering an elastography technique based on the Yeoh hyperelastic model. The strength of the Yeoh model has been validated through compression testing of breast phantoms (small and large sizes), animal tissues, and in-vivo human tissues. The proposed method provides thresholds for the mechanical properties of soft tissues, which are useful in medical applications.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Farid Golnaraghi
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.

A Dental Assisting System for Procedures Performed by Air–Turbine Handpieces

Author: 
Date created: 
2013-08-13
Abstract: 

The present thesis introduces a dental assisting system (DAS) for procedures that are performed by air–turbine dental handpieces. Dental restoration is a process that begins with removing carries and affected tissues to retain the functionality of tooth structures. Air–turbine dental handpieces are high–speed rotary cutting tools that are widely used by dentists during this operation. The next stage in the process is filling the cavity with appropriate restorative materials. “Amalgam” and “composite” are two dental restorative materials that are extensively used by dentists. Most old restorations eventually fail and need to be replaced. One of the difficulties in replacing failing restorations is discerning the boundary of the restorative materials. Dentists may remove healthy tooth structures while replacing tooth–colored composites. Although the visibility issue is less challenging for amalgam materials, replacing them still results in loss of healthy tooth layers. Developing an objective and sensor–based method is a promising approach to monitor restorative operations and prevent removal of healthy tooth structures. The designed DAS uses the audio signals of ATDH during the cutting process. Audio signals are rich sources of information and can be analysed to identify a particular zone of cutting. Support vector machine (SVM), a powerful algorithm for classification, is employed to differentiate the tooth structures from composite/amalgam samples based on their cutting sounds. The averaged short–time Fourier transform coefficients are selected as the features; and the performance of the SVM classifier is evaluated from different aspects such as number of features, feature scaling methods, and the utilized kernels. The obtained results indicated capability and efficiency of the proposed scheme. The developed DAS can also measure the speed of ATDH, and maintain it during loaded conditions. An indirect speed measurement method is introduced based on the vibration/sound of ATDH. This measurement technique is explained theoretically based on the rotating unbalance concept and the vibration of a fixed–free beam. To control the speed, a proportional–integral controller is designed and tested. The feasibility of this controller in maintaining the speed in the loaded conditions was confirmed by simulations and experiments.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Siamak Arzanpour
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) Ph.D.

Implementation of EKF-SLAM on NAO Humanoid Robot

Date created: 
2013-07-08
Abstract: 

Autonomous navigation is among the most important areas within the robotic community and has gained significant interest in the last three decades. At its core, autonomous navigation depends on whether or not the Simultaneous Localization And Mapping (SLAM) problem is properly addressed. SLAM is the mobile robot’s ability to navigate in an unknown (indoor or outdoor) environment. The process includes building a map of the environment while concurrently the mobile robot position is being updated. This thesis focuses on the implementation of SLAM on NAO humanoid robot. The ubiquitous method of Extended Kalman Filter (EKF) is employed in this project. The central focus of the thesis is on simulation and the real time implementation of EKF-SLAM with the entire coding in python. I have adopted the feature-based map (landmarks) along with updating the robot position in an indoor area. The main contribution of this work is the real time implementation of EKF-SLAM on the NAO humanoid robot, which is manufactured by Aldebaran Robotics. The NAO humanoid robot has been programmed in python through the interfacing with the main software of NAO that is called NAOqi. The details of implementing EKF-SLAM on NAO are explained and discussed in this thesis. The main two sensors used for addressing SLAM on the NAO humanoid robot were odometry and laser sensors. In addition, this thesis has practically solved the data association problem, which is considered one of the main issues of EKF-SLAM, using Mahalanobis Distance likelihood technique. Data association is the ability of the robot to correspond the detected landmarks from sensors at any time with the pre-observed landmarks in the map. Throughout the experimental studies, I have detected and successfully resolved several problems. First, the deviation in NAO robot motion function is solved by designing a closed loop motion control of the robot. Second, the Agglomerative Hierarchical Clustering (AHC) method is used to let the laser sensor detects and differentiates between different landmarks at one time. The laser in NAO generates the world around it via 683 points without differentiation between different objects in the scene. Third, obstacle avoidance function is designed.

Document type: 
Thesis
File(s): 
Senior supervisor: 
Ahmad Rad
Department: 
Applied Sciences: School of Mechatronic Systems Engineering
Thesis type: 
(Thesis) M.A.Sc.