Author: BORDALBA LLABERIA, RICARD
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN AUTOMATIC CONTROL, ROBOTICS AND VISION
Department: Institute of Robotics and Industrial Informatics (IRI)
Mode: Temporary seizure
Deposit date: 28/07/2021
Reading date: pending
Reading time: pending
Reading place: pending
Thesis director: ROS GIRALT, LLUIS | PORTA PLEITE, JOSE MARIA
Committee:
     PRESIDENT: GONZÁLEZ DE SANTOS, PABLO
     SECRETARI: THOMAS ARROYO, FEDERICO
     VOCAL: LAVALLE, STEVEN
Thesis abstract: This work proposes a methodology for kinodynamic planning and trajectory control in robots with closed kinematicchains. The ability to plan trajectories is key in a robotic system, as it provides a means to convert high-level taskcommands¾like ¿move to that location'', or ¿throw the object at such a speed''¾into low-level controls to befollowed by the actuators. In contrast to purely kinematic planners, which only generate collision-free paths inconfiguration space, kinodynamic planners compute state-space trajectories that also account for the dynamics andforce limits of the robot. In doing so, the resulting motions are more realistic and exploit gravity, inertia, andcentripetal forces to the benefit of the task. Existing kinodynamic planners are fairly general and can deal withcomplex problems, but they require the state coordinates to be independent. Therefore, they are hard to apply torobots with loop-closure constraints whose state space is not globally parameterizable. These constraints define anonlinear manifold on which the trajectories must be confined, and they appear in many systems, like parallelrobots, cooperative arms manipulating an object, or systems that keep multiple contacts with the environment. Inthis work, we propose three steps to generate optimal trajectories for such systems. In a first step, we determine atrajectory that avoids the collisions with obstacles and satisfies all kinodynamic constraints of the robot, includingloop-closure constraints, the equations of motion, or any limits on the velocities or on the motor and constraintforces. This is achieved with a sampling-based planner that constructs local charts of the state space numerically, and with an efficient steering method based on linear quadratic regulators. In a second step, the trajectory isoptimized according to a cost function of interest. To this end we introduce two new collocation methods fortrajectory optimization. While current methods easily violate the kinematic constraints, those we propose satisfythese constraints along the obtained trajectories. During the execution of a task, however, the trajectory may beaffected by unforeseen disturbances or model errors. That is why, in a third step, we propose two trajectory controlmethods for closed-chain robots. The first method enjoys global stability, but it can only control trajectories thatavoid forward singularities. The second method, in contrast, has local stability, but allows these singularities to betraversed robustly. The combination of these three steps expands the range of systems in which motion planningcan be successfully applied.

Author: GÓMEZ KABELKA, PAU
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN PHOTONICS
Department: Institute of Photonic Sciences (ICFO)
Mode: Normal
Deposit date: 06/07/2021
Reading date: 28/09/2021
Reading time: 10:00
Reading place: ICFO - The Institute of Photonic Sciences. Av. Carl Friedrich Gauss, 3
Thesis director: MITCHELL, MORGAN
Committee:
     PRESIDENT: HIRANO, TAKUYA
     SECRETARI: TARRUELL, LETICIA
     VOCAL: KLEMPT, CARSTEN
Thesis abstract: In this work we demonstrate the first realization of a comagnetormeter in the ultracold regime. In comparison to regular magnetometers, which are designed to maximize their magnetic field sensitivity, a comagnetometer uses paired magnetometers in a differential configuration to cancel the effects of the magnetic field and resolve weak dynamics that differently affect its constituents. Here, we implement a omagnetometer within the f=1 and f=2 ground state hyperfine manifolds of a Rb87 spinorBose-Einstein condensate (SBEC). The hyperfine manifolds feature nearly opposite gyromagnetic ratios and thus the sum of theirprecession angles is only weakly coupled to external magnetic fields, while being highly sensitive to any effect that rotates bothmanifolds in the same way.A fundamental limitation of the comagnetometer is f=2--> f=1 hyperfine relaxing collisions, where the liberated kinetic energy expels colliding atoms from the optical trap. These collisions are state-dependent and can be avoided by preserving the f=2 spin state in a stretched configuration. We show how this can be achieved at low magnetic fields, where the spin-dependent contact interaction is the dominant energy contribution and stabilizes the spin orientation of the SBEC. Under these conditions, the comagnetometer coherence time can be extended to ~1 s and the observed common magnetic field suppression is 44.0(8)dB. The technique is applied to precision measurement of the interhyperfine interaction in 87Rb. The uncertainty in the obtained interhyperfine scattering lengths is reduced by more than a factor three with respect to previously reported values. We also present preliminary studies on phase-resolved parametric amplification within a SBEC comagnetometer. In this case, the f=2 manifold undergoes parametric amplification, while the f=1 manifold keeps track of the rotating reference frame induced by the applied external magnetic field.We describe technical improvements to the experimental system in two areas: magnetic control and manipulation, and opticaltrapping and probing. The first group of improvements includes the implementation of radiofrequency (rf) and microwave (mw)driving and the development of a real-time rf source. The second group of improvements includes a pulsed optical trapping technique, a digital implementation of the laser locking scheme, and a hyperfine-selective Faraday probing method.

Author: MAHAJAN, ANKUSH
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN SIGNAL THEORY AND COMMUNICATIONS
Department: Department of Signal Theory and Communications (TSC)
Mode: Normal
Deposit date: 27/07/2021
Reading date: 28/09/2021
Reading time: 11:00
Reading place: Sala d'Actes - EETAC - Campus Castelldefels - meet.google.com/uyo-reyh-haa
Thesis director: MARTINEZ RIVERA, RICARDO VICTOR | CHRISTODOULOPOULOS, KONSTANTINOS | MUÑOZ GONZALEZ, RAUL
Committee:
     PRESIDENT: KIRSTADTER, ANDREAS
     SECRETARI: COMELLAS COLOME, JAUME
     VOCAL NO PRESENCIAL: TORNATORE, MASSIMO
Thesis abstract: The tremendous increase in data traffic has spurred a rapid evolution of the optical networks for a reliable, affordable, cost effective and scalable network infrastructure. To meet some of these requirements, network operators are pushing toward disaggregation. Network disaggregation focuses on decoupling the traditional monolithic optical transport hardware into independent functional blocks that interoperate. This enables a relatively free market where the network operators/owners could choose the best-in-class equipment from different vendors overcoming the vendor lock-in, at better prices. In this multi-vendor disaggregation context, the used equipment would impact the physical layer and the overall network behavior. This results in increasing the uncertainty on the performance when compared to a traditional single vendor aggregated approach.For effective optical network planning, operation and optimization, it is necessary to estimate the Quality of Transmission (QoT) of the connections. Network designers are interested in accurate and fast QoT estimation for services to be established in a future or existing network. Typically, QoT estimation is performed using a Physical Layer Model (PLM) which is included in the QoT estimation tool or Qtool. A design margin is generally included in a Qtool to account for the modeling and parameter inaccuracies, to re-assure an acceptable performance. PLM accuracy is highly important as modeling errors translate into a higher design margin which in turn translate into wasted capacity or unwanted regeneration. Recently monitoring and machine learning (ML) techniques have been proposed to account for the actual network conditions and improving the accuracy of the PLM in single vendor networks. This in turn results in more accurate QoT estimation. The first part of the thesis focuses on the ML assisted accurate QoT estimation techniques. In this regard, we developed a model that uses monitoring information from an operating network combined with supervised ML regression techniques to understand the network conditions. In particular, we model the generated penalties due to i). EDFA gain ripple effect, and ii). filter spectral shape uncertainties at ROADM nodes. Furthermore, with the aim of improving the Qtool estimation accuracy in multi-vendor networks, we propose PLM extensions. In particular, we introduce four TP vendor dependent performance factors that capture the performance variations of multi-vendor TPs. To verify the potential improvement, we studied the following two use cases with the proposed PLM, to: i) optimize the transponders (TPs) launch power; and ii) reduce design margin in incremental planning. In consequence, the last part of this thesis aims at investigating and solving the issue of accuracy limitation of Qtool in dynamic optimization tasks. To keep the models aligned to the real conditions, the digital twin (DT) concept is gaining significant attention in the research community. The DT is more than a model of the system; it includes an evolving set of data, and a means to dynamically adjust the model. Based on the DT fundamentals, we devised and implemented an iterative closed control loop process that, after several intermediate iterations of the optimization algorithm, configures the network, monitors, and retrains the Qtool. For the Qtool retraining, we adopt a ML-based nonlinear regression fitting technique. The key advantage of this novel scheme is that whilst the network operates, the Qtool parameters are retrained according to the monitored information with the adopted ML model. Hence, the Qtool tracks the projected states intermediately calculated by the algorithm. This reduces the optimization time as opposed to directly probing and monitoring the network.

Author: DEUTSCHMANN, INA MARIA
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN APPLIED MATHEMATICS
Department: Department of Mathematics (MAT)
Mode: Normal
Deposit date: 27/07/2021
Reading date: 29/09/2021
Reading time: 13:30
Reading place: FME - UPC CAMPUS SUD - VIDEOCONFERÈNCIA :https://meet.google.com/xom-nyse-ewg
Thesis director: LOGARES MAURIE, RAMIRO
Committee:
     PRESIDENT: BITTNER, LUCIE
     SECRETARI: CASANELLAS RIUS, MARTA
     VOCAL: LOPEZ, PHILIPPE
Thesis abstract: There is a myriad of microorganisms on Earth contributing to global biogeochemical cycles, and their interactions are considered pivotal for ecosystem function. Previous studies have already determined relationships between a limited number of microorganisms. Yet, we still need to understand a large number of interactions to increase our knowledge of complex microbiomes. This is challenging because of the vast number of possible interactions. Thus, microbial interactions still remain barely known to date. Networks are a great tool to handle the vast number of microorganisms and their connections, explore potential microbial interactions, and elucidate patterns of microbial ecosystems. This thesis locates at the intersection of network inference and network analysis. The presented methodology aims to support and advance marine microbial investigations by reducing noise and elucidating patterns in inferred association networks for subsequent biological down-stream analyses. This thesis¿s main contribution to marine microbial interactions studies is the development of the program EnDED (Environmentally-Driven Edge Detection), a computational framework to identify environmentally-driven associations inside microbial association networks, inferred from omics datasets. We applied the methodology to a model marine microbial ecosystem at the Blanes Bay Microbial Observatory (BBMO) in the North-Western Mediterranean Sea (ten years of monthly sampling). We also applied the methodology to a dataset compilation covering six global-ocean regions from the surface (3 m) to the deep ocean (down to 4539 m). Thus, our methodology provided a step towards studying the marine microbial distribution in space via the horizontal (ocean regions) and vertical (water column) axes.

Author: MARTÍNEZ-DENEGRÍ SÁNCHEZ, GUILLERMO
Thesis file: (contact the Doctoral School to confirm you have a valid doctoral degree and to get the link to the thesis)
Programme: DOCTORAL DEGREE IN PHOTONICS
Department: Institute of Photonic Sciences (ICFO)
Mode: Normal
Deposit date: 23/07/2021
Reading date: 29/09/2021
Reading time: 11:00
Reading place: ICFO ¿ The Institute of Photonic Sciences - Campus Baix Llobregat - Av. Carl Friedrich Gauss, 3 08860 Castelldefels (Barcelona) - SPAIN
Thesis director: MARTORELL PENA, JORDI
Committee:
     PRESIDENT: MORANTE LLEONART, JUAN RAMÓN
     SECRETARI: ROYO ROYO, SANTIAGO
     VOCAL: FABREGAT SANTIAGO, FRANCISCO
Thesis abstract: The environmental issues associated with the use of conventional fuels necessitates the utilisation of renewable energy sources, as well as the implementation of energy efficient designs, in order to decrease electricity consumption. Photovoltaic (PV) technology can be employed for both approaches by converting not only natural but, also, artificial light into electricity. Among the different emerging PVs, perovskites achieve the highest power conversion efficiency, providing a widely tuneable bandgap with minimum open circuit losses. Moreover, their fabrication uses readily available materials, and does not necessarily require either the use of high temperature processes or vacuum deposition techniques. In this thesis, we enhance light harvesting in perovskite solar cells, and approach the energy efficiency concept through their optimised fabrication and integration in light selective structures. This is accomplished by the implementation of optical and material strategies applied to specific perovskite solar cell designs. The results prove that such strategies provide enhanced light absorption and optimal PV performance in low temperature devices, and enable the recycling of light into electricity for alternative photonic applications. The approaches presented could be utilised in future procedures to decrease the amount of Pb employed in perovskite solar cells, and to reduce the energy consumption during fabrication and the operation of other optoelectronic devices.The thesis is organised into four chapters. Chapter 1 serves as an introduction, where the current energy situation and PV technology are analysed, together with an insight into light harvesting and energy efficiency in perovskite solar cells. In Chapter 2, we demonstrate the employment of a periodic structure to propagate ergodic light in order to increase light absorption in perovskite solar cells, as would happen by employing randomly textured surfaces. This structure serves as a tool to decrease the Pb content used in perovskite solar cells, since 30% less material can be used to obtain a solar cell with equal performance. Then, in Chapter 3, the same periodic configuration with a thin film structure deposited on its surface is applied as a waveguide, which is also able to transmit polarised light. Moreover, two perovskite solar cells integrated on the sides recycle the non-transmitted light into electricity, increasing the energy efficiency of the optical process, with further application in liquid crystal displays (LCDs). Finally, in Chapter 4, we demonstrate the suitable application of a nanoparticle bilayer made of one layer of SnO2 and another of TiO2 as n-type materials in perovskite solar cells. These types of devices, based on low temperature processes, are proven to perform better than those containing one type of nanoparticles, especially in semi-transparent devices. In such devices we achieved an enhancement in performance of up to 30% for solar cells based on extremely thin active layers.