The version posted must include the following notice on the first page: This is the peer reviewed version of the following article: http://onlinelibrary.wiley.com/doi/10.1002/cne.23477/abstract, which has been published in final form at DOI: 10.1002/cne.23477. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving. / 京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第19610号 / 医博第4117号 / 新制||医||1015(附属図書館) / 32646 / 京都大学大学院医学研究科医学専攻 / (主査)教授 渡邉 大, 教授 髙橋 良輔, 教授 宮本 享 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

This thesis considers the problem of using signals of opportunity (SOO) with known direction of arrival (DOA) for aircraft positioning. SOO is a collective name for a wide range of signals not intended for navigation but which can be intercepted by the radar warning system on an aircraft. These signals can for example aid an unassisted inertial navigation system (INS) in areas where the global navigation satellite system (GNSS) is inaccessible. Challenges arise as the signals are transmitted from non-controllable sources without any guarantee of quality and availability. Hence, it is important that any estimation method utilising SOO is robust and statistically consistent in case of time-varying signals of different quality, missed detections and unreliable signals such as outliers. The problem is studied using SOO sources with either known or unknown locations. An extended Kalman filter (EKF) based solution is proposed for the first case which is shown to significantly improve the localisation performance compared to an unassisted INS in common scenarios. Yet, a number of factors affect this performance, including the measurement noise variance, the signal rate and the availability of known source locations. An outlier rejection mechanism is developed which is shown to increase the robustness of the suggested method. A numerical evaluation indicates that statistical consistency can be maintained in many situations even with the above-mentioned challenges. An EKF based simultaneous localisation and mapping (SLAM) solution is proposed for the case with unknown SOO source locations. The flight trajectory and initialisation process of new SOO sources are critical in this case. A method based on nonlinear least squares is proposed for the initialisation process, where new SOO sources are only allowed to be initialised in the filter once a set of requirements are fulfilled. This method has shown to increase the robustness during initialisation, when the outlier rejection is not applicable. When combining known and unknown SOO source locations, a more stable localisation solution is obtained compared to when all locations are unknown. Applicability of the proposed solution is verified by a numerical evaluation. The computational time increases cubically with the number of sources in the state and quadratically with the number of measurements. The time is substantially increased during landmark initialisation.

This thesis investigates the important potential of Cooperative Intelligent Transport Systems (C-ITS) onmobility and society. C-ITS appears as a promising solution to reinvent transportation and become avital part of the ever-evolving environment as developments in technology continue to change the world.The goal of this study is to investigate how C-ITS can enhance and promote various forms of mobility.It additionally investigates at how C-ITS applications and policy objectives interact, highlighting C-ITS’contribution to the development of a sustainable society.A thorough examination of the current literature, case studies, and pertinent policies was conductedin order to analyse the possible advantages and difficulties related to C-ITS in detail. In order toprovide seamless communication and interaction between C-ITS systems and other devices, the researchemphasizes the importance of early integration and adoption of C-ITS as a solution. It also emphasizesthe need for standardization, interoperability, and collaborative efforts among stakeholders.Findings demonstrate that C-ITS has the capacity to support policies aimed at improving transportationsystems and mobility in the cities. C-ITS usage has enormous potential for influencing society andmobility. C-ITS reduces accidents while enhancing road safety through real-time communication. Byenhancing traffic flow and promoting alternative modes of transportation, it supports environmentalsustainability. It also has secondary effects such as reducing pollutants and improving air andnoise quality. Through the integration of numerous mobility choices and the provision of real-timeinformation, C-ITS improves accessibility. For implementation to be successful, privacy and securityissues as well as economic reasons must be taken into consideration. To solve the issues posed byconcerns about data privacy, security, and economic factors, however, strong policies, legislation,and safe data processing techniques are needed. C-ITS has the potential to help create a future oftransportation that is safer, more environmentally friendly, and more effective.In the concluding part, the paper suggests numerous possibilities for C-ITS research going forward.It advises combining policies and guiding documents to offer a clearer strategy for utilizing C-ITSsuccessfully. Additionally, creating more complex mathematical models that include equations can helpus comprehend the importance of the variables better. Iterative procedures integrated into detailedmodels allow for the comparison of many scenarios, addressing the various desires of stakeholders andexperts. Additionally, combining C-ITS with Vehicle-to-Everything (VoT) systems offer a chance toinvestigate the real advantages and make it simpler to make comparisons with other variables. Furtherresearch should be carried out on the likelihood of developing an automated mobility system.

Concepts of tree coding and of rate-distortion theory are applied to the problem of the transmission of analog signals over digital channels. Coding schemes are developed which yield improvements of up to six dB in signal-to-noise ratio over conventional techniques for the reproduction of speech waveforms. / Thesis / Master of Engineering (MEngr)

Drilling tool wear can significantly affect the performance of the drilling operation and add extra cost to it. Accurate detection of drilling tool condition is very important for enabling proactive maintenance, minimizing downtime, and optimizing drilling processes.  This study investigates the possibility of detecting drilling tool condition of a Raisboring machine using drilling signals with deep learning methods. Given the current situation where the operators of the machine are responsible for detecting drilling abnormalities, which introduces bias and inconsistency to the process, it is crucial to develop an automated machine health monitoring system.  The objectives of this study were to explore the effectiveness of deep learning approaches in detecting drilling tool faults based on sensor data collected during drilling operations; as well as to find out which drilling signal is most effective for this problem.  The working dataset consists of labeled samples representing two drilling tool conditions (new and worn) and includes four channels: RPM, torque, feed force, and ground acceleration signals. To implement this, both time-domain features and frequency-domain features were extracted from the drilling signals and used as input to fully connected neural networks (FCNNs) and convolutional neural networks (CNNs). Performance metrics such as accuracy, precision, recall, and F1-score were used to assess the models’ performance. The results indicate that deep learning has great potential in detecting drilling tool condition. More specifically, the vibration signal, which yielded the highest results with the different algorithms.  The study highlights the potential of deep learning techniques for real-time, automated monitoring of drilling tool condition, enabling timely maintenance interventions and enhanced operationalefficiency.

No description available.

No description available.

Husqvarna AB is one of the leading producers of outdoor products such as autonomous lawn mowers. One important feature of these products is the ability toquickly respond to environmental factors such as slippy areas. A reliable slip detector is needed for this mission and many different technologies exists for detectingslip events. A common technique is to check the wheel motor current, which clearlydeviates when the lawn mower is subjected to slipping. The on-board sensors opensup for an alternative solution which utilizes the loop sensors as the main slip detector. This thesis covers the construction of a slip detection prototype which is basedon the loop sensors. In the end, Husqvarna AB was provided with a new alternativesolution, which was successfully compared to the exiting solution. It proved to bea reliable slip detector for manually induced slipping indoors, outdoor performancewere not investigated. Ultimately, the implemented prototype outperformed the existing solution in the intended environment of indoor testing.

Standards and regulations for dimensioning of load-bearing structures are based on the response of load-bearing structures subjected to loads without variation in time. In the event of an accidental load, e.g. a collision or explosion, causes the load to have a rapid variation in the time resulting in a dynamic response. Previous studies have shown that structures that respond in a certain way under static load have shown a completely different behavior under dynamic influence and therefore it is of interest to study the dynamic response of structures.By testing concrete beams with varying amounts of shear reinforcement subjected to impact loading, the purpose of this report is to analyze how the beams responds in terms of crack width and vibrations when they are exposed to a dynamic load. For comparison, reference tests have also been performed on beams subjected to a quasi-static load.The result of the project shows that the shear reinforcement comes into play as the beams with a larger amount of reinforcement have more capacity to hold the flexural shear cracks together. There is also an indication that the dynamic flexural shear capacity could be lower than static shear capacity as the shear cracks had an increased inclination during dynamic loading for some of the beams. This results in a decreased flexural shear capacity as only one stirrup carried the load across the shear crack.Measured signal shows that beams failing respond when impacted by the similar to a plastic collision, while beams responding with a flexure dominated mode without going to failure instead answer similar to an elastic collision. Furthermore, there is indication that the natural frequencies change significantly due to both flexural cracks and flexural shear cracks.

Ultrasound imaging is a safe and popular imaging technique that relies on received radio frequency (RF) echos to show the internal organs and tissue. B-mode (Brightness mode) is the typical mode of ultrasound images generated from RF signals. In practice, the real processing algorithms from RF signals to B-mode images in ultrasound machines are kept confidential by the manufacturers. The thesis aims to estimate the process and reproduce the same results as the Ultrasonix One ultrasound machine does using deep learning. 11 scalar parameters including global gain, time-gain-compensation (TGC1-8), dynamic range and reject affect the transformation from RF signals to B-mode images in the machine. Data generation strategy was proposed. Two network architectures adapted from U-Net and Tiramisu Net were investigated and compared. Results show that a deep learning network is able to translate RF signals to B-mode images with respect to the controlling parameters. The best performance is achieved by adapted U-Net that reduces per pixel error to 1.325%. The trained model can be used to generate images for other experiments.