Motor Unit Spike Trains Recontruction from sEMG Signals for Gesture Classification on a Low-Power Processing Platform

Hand gesture recognition based on surface electromyography (sEMG) signals is a promising approach for the development of intuitive human-machine interfaces (HMIs) in domains such as robotics and prosthetics. The sEMG signal arises from the muscles' electrical activity, and can thus be used to recognize hand gestures. The decoding from sEMG signals to actual control signals is non-trivial; typically, control systems map sEMG patterns into a set of gestures using machine learning, failing to incorporate any physiological insight. This master thesis aims at developing a bio-inspired hand gesture recognition system based on neuromuscular spike extraction rather than on simple pattern recognition. The system relies on a decomposition algorithm based on independent component analysis (ICA) that decomposes the sEMG signal into its constituent motor unit spike trains, which are then forwarded to a machine learning classifier. Since ICA does not guarantee a consistent motor unit ordering across different sessions, 3 approaches are proposed: 2 ordering criteria based on firing rate and negative entropy, and a re-calibration approach that allows the decomposition model to retain information about previous sessions. Using a multilayer perceptron (MLP), the latter approach results in an accuracy up to 99.4% in a 1-subject, 1-degree of freedom scenario. Afterwards, the decomposition and classification pipeline for inference is parallelized and profiled on the PULP platform, achieving a latency < 50 ms and an energy consumption < 1 mJ. Both the classification models tested (a support vector machine and a lightweight MLP) yielded an accuracy > 92% in a 1-subject, 5-classes (4 gestures and rest) scenario. These results prove that the proposed system is suitable for real-time execution on embedded platforms and also capable of matching the accuracy of state-of-the-art approaches, while also giving some physiological insight on the neuromuscular spikes underlying the sEMG.

Effetti dell’arousal sul fitness to drive: algoritmi e simulazione

Diversi elementi possono influenzare il comportamento dei conducenti mentre svolgono attività di guida. Dalle distrazioni visive a quelle cognitive, le emozioni (che potrebbero risultare da dati biometrici, come temperatura, battito cardiaco, pressione, ecc.) e le condizioni di altri conducenti, che possono svolgere un ruolo significativo, fungendo da fattore che può aumentare il tempo di risposta dei guidatori. Mantenere sotto controllo questi parametri potrebbe essere fondamentale per evitare situazioni pericolose, per decidere ed eseguire azioni che potrebbero influenzare il verificarsi di incidenti stradali. Questo volume analizza l'indice "Fitness-to-Drive" e mira a valutare come gli effetti dell'eccitazione possono influenzare lo stato dei conducenti. La tesi presenta alcune valutazioni sperimentali condotte su un simulatore di guida, discutendo i risultati ottenuti.

A study on tackling visual odometry by a transformer architecture

This dissertation describes a deepening study about Visual Odometry problem tackled with transformer architectures. The existing VO algorithms are based on heavily hand-crafted features and are not able to generalize well to new environments. To train them, we need carefully fine-tune the hyper-parameters and the network architecture. We propose to tackle the VO problem with transformer because it is a general-purpose architecture and because it was designed to transformer sequences of data from a domain to another one, which is the case of the VO problem. Our first goal is to create synthetic dataset using BlenderProc2 framework to mitigate the problem of the dataset scarcity. The second goal is to tackle the VO problem by using different versions of the transformer architecture, which will be pre-trained on the synthetic dataset and fine-tuned on the real dataset, KITTI dataset. Our approach is defined as follows: we use a feature-extractor to extract features embeddings from a sequence of images, then we feed this sequence of embeddings to the transformer architecture, finally, an MLP is used to predict the sequence of camera poses.

Mitigating non-Lambertian surfaces issues in Stereo Matching with Neural Radiance Fields

Depth estimation from images has long been regarded as a preferable alternative compared to expensive and intrusive active sensors, such as LiDAR and ToF. The topic has attracted the attention of an increasingly wide audience thanks to the great amount of application domains, such as autonomous driving, robotic navigation and 3D reconstruction. Among the various techniques employed for depth estimation, stereo matching is one of the most widespread, owing to its robustness, speed and simplicity in setup. Recent developments has been aided by the abundance of annotated stereo images, which granted to deep learning the opportunity to thrive in a research area where deep networks can reach state-of-the-art sub-pixel precision in most cases. Despite the recent findings, stereo matching still begets many open challenges, two among them being finding pixel correspondences in presence of objects that exhibits a non-Lambertian behaviour and processing high-resolution images. Recently, a novel dataset named Booster, which contains high-resolution stereo pairs featuring a large collection of labeled non-Lambertian objects, has been released. The work shown that training state-of-the-art deep neural network on such data improves the generalization capabilities of these networks also in presence of non-Lambertian surfaces. Regardless being a further step to tackle the aforementioned challenge, Booster includes a rather small number of annotated images, and thus cannot satisfy the intensive training requirements of deep learning. This thesis work aims to investigate novel view synthesis techniques to augment the Booster dataset, with ultimate goal of improving stereo matching reliability in presence of high-resolution images that displays non-Lambertian surfaces.

Ephemeris generation of the Galilean moons by using VLBI observations in support to stellar occultations

Stellar occultations are the most accurate Earth-based astronomy technique to obtain the lateral position of celestial bodies, in the case of natural satellites, their accuracy also depends on the central body to which the satellite orbits. The main goal of this thesis work is to analyze if and how very long baseline interferometry (VLBI) measurements of a body like Jupiter can be used in support to stellar occultations of its natural satellites by reducing the planetary uncertainty at the time of the occultation. In particular, we analyzed the events of the stellar occultations of Callisto (15.01.2024) and Io (02.04.2021). The stellar occultation of Callisto has been predicted and simulated using the stellar occultation reduction analysis (SORA) toolkit while the stellar occultation of Io has been already studied by Morgado et al. We then simulated the VLBI data of Jupiter according to the current JUNO trajectories. The required observation were then used as input of an estimation to which then we performed a covariance analysis on the estimated parameters to retrieve the formal errors (1 − σ uncertainties) at each epoch of the propagation. The results show that the addition of the VLBI slightly improves the uncertainty of Callisto even when Jupiter knowledge is worse while for Io we observed that the VLBI data is especially crucial in the scenario of an a priori uncertainty in Jupiter state of about 10km. Here we can have improvements of the estimated initial states of Io of about 70m, 230m and 900m to the radial, along-track and cross-track directions respectively. Moreover, we have also obtained the propagated errors of the two moons in terms of right ascension and declination which both show uncertainties in the mas level at the occultation time. Finally, we simulated Io and Europa together and we observed that at the time of the stellar occultation of Europa the along-track component of Io is constrained, confirming the coupling between the two inner moons.

An introduction to scalar-tensor gravity: theory and experiment in the Brans-Dicke model

Le teorie della gravità scalare-tensore sono una classe di teorie alternative alla rel- atività generale in cui l’interazione gravitazionale è descritta sia dalla metrica, sia da un campo scalare. Ne costituisce un esempio caratteristico la teoria di Brans-Dicke, in- trodotta come estensione della relatività generale in modo da renderla conforme con il principio di Mach. Il presente lavoro di tesi è volto a presentare un’analisi di questa teoria nei suoi aspetti principali, studiandone i fondamenti teorici e il modello cosmologico derivante, sottolineandone inoltre i limiti e le criticità; in seguito vengono esposti i risultati degli esperimenti fino ad ora svolti per verificare fondamenti e previsioni del modello.

L’efficacia del protocollo di Askling nel return to play del calciatore post lesione muscolare agli Hamstring: revisione sistematica della letteratura

Background: Gli infortuni muscolari generano grande preoccupazione nella medicina sportiva e costituiscono la più frequente causa di assenza dal campo nel calcio. Nello specifico , a causa dei gesti tecnici caratteristici di questo sport, le HSI (lesioni muscolari agli Hamstring) sono le più frequenti nei calciatori, costringendoli all’assenza dal campo più di qualsiasi altro infortunio. Il protocollo di Askling è uno degli studi più popolari e riconosciuti nella riabilitazione delle HSI. Obiettivi: L’obiettivo della revisione sistematica è valutare l’efficacia del protocollo di Askling nel RTP del calciatore post lesione muscolare agli Hamstring. Questo contributo, quindi, discuterà gli attuali concetti clinici riguardanti la riabilitazione delle HSI, analizzando trial clinici randomizzati che confrontino gli effetti del protocollo di Askling con altri protocolli.

 Materiali e Metodi: Nella selezione degli articoli, sono state indagate le seguenti banche dati: “PubMed”, “PEDro” e “Cochrane Library”. Una volta realizzato il PICOS, questo è stato convertito nelle parole chiave pertinenti all’argomento di interesse della revisione. Sono stati presi in esame unicamente studi RCTs, di qualsiasi anno di pubblicazione e lingua, reperibili in modalità Full Text attraverso il servizio proxy offerto dall’Università di Bologna. Gli studi per essere inclusi devono avere un punteggio PEDro score > 5. Risultati: Il tempo medio del Return To Play è stato ridotto di 23 giorni, da 51 (C-Protocol) a 28 giorni (L-Protocol). Non vi è però alcuna differenza significativa tra i gruppi per i tassi di re-injury entro 2 mesi, da 2 a 6 mesi e da 6 a 12 mesi.
 Conclusioni: Emerge dalla letteratura che un protocollo riabilitativo costituito principalmente da esercizi che prediligono l’allungamento muscolare risulta più efficace di un protocollo convenzionale nel ridurre il tempo trascorso dall'infortunio all'allenamento completo senza restrizioni e/o al match play.

Image Processing Algorithms applied to Radio SLAM using MIMO Radars

Radio Simultaneous Location and Mapping (SLAM) consists of the simultaneous tracking of the target and estimation of the surrounding environment, to build a map and estimate the target movements within it. It is an increasingly exploited technique for automotive applications, in order to improve the localization of obstacles and the target relative movement with respect to them, for emergency situations, for example when it is necessary to explore (with a drone or a robot) environments with a limited visibility, or for personal radar applications, thanks to its versatility and cheapness. Until today, these systems were based on light detection and ranging (lidar) or visual cameras, high-accuracy and expensive approaches that are limited to specific environments and weather conditions. Instead, in case of smoke, fog or simply darkness, radar-based systems can operate exactly in the same way. In this thesis activity, the Fourier-Mellin algorithm is analyzed and implemented, to verify the applicability to Radio SLAM, in which the radar frames can be treated as images and the radar motion between consecutive frames can be covered with registration. Furthermore, a simplified version of that algorithm is proposed, in order to solve the problems of the Fourier-Mellin algorithm when working with real radar images and improve the performance. The INRAS RBK2, a MIMO 2x16 mmWave radar, is used for experimental acquisitions, consisting of multiple tests performed in Lab-E of the Cesena Campus, University of Bologna. The different performances of Fourier-Mellin and its simplified version are compared also with the MatchScan algorithm, a classic algorithm for SLAM systems.

Virtualization of Wiring Harness Manipulation Tasks through the PyChrono Simulation Engine

The purpose of this thesis work is the study and creation of a harness modelling system. The model needs to simulate faithfully the physical behaviour of the harness, without any instability or incorrect movements. Since there are various simulation engines that try to model wiring's systems, this thesis work focused on the creation and test of a 3D environment with wiring and other objects through the PyChrono Simulation Engine. Fine-tuning of the simulation parameters were done during the test to achieve the most stable and correct simulation possible, but tests showed the intrinsic limits of the Engine regarding the collisions' detection between the various part of the cables, while collisions between cables and other physical objects such as pavement, walls and others are well managed by the simulator. Finally, the main purpose of the model is to be used to train Artificial Intelligence through Reinforcement Learnings techniques, so we designed, using OpenAI Gym APIs, the general structure of the learning environment, defining its basic functions and an initial framework.

Jeans modeling of massive early-type galaxies with observed stellar kinematic profiles up to 2-3 effective radii

A recent integral-field spectroscopic (IFS) survey, the MASSIVE survey (Ma et al. 2014), observed the 116 most massive (MK < −25.3 mag, stellar mass M∗ > 10^11.6 M⊙) early-type galaxies (ETGs) within 108 Mpc, out to radii as large as 40 kpc, that correspond to ∼ 2 − 3 effective radii (Re). One of the major findings of the MASSIVE survey is that the galaxy sample is split nearly equally among three groups showing three different velocity dispersion profiles σ(R) outer of a radius ∼ 5 kpc (falling, flat and rising with radius). The purpose of this thesis is to model the kinematic profiles of six ETGs included in the MASSIVE survey and representative of the three observed σ(R) shapes, with the aim of investigating their dynamical structure. Models for the chosen galaxies are built using the numerical code JASMINE (Posacki, Pellegrini, and Ciotti 2013). The code produces models of axisymmetric galaxies, based on the solution of the Jeans equations for a multicomponent gravitational potential (supermassive black hole, stars and dark matter halo). With the aim of having a good agreement between the kinematics obtained from the Jeans equations, and the observed σ and rotation velocity V of MASSIVE (Veale et al. 2016, 2018), I derived constraints on the dark matter distribution and orbital anisotropy. This work suggests a trend of the dark matter amount and distribution with the shape of the velocity dispersion profiles in the outer regions: the models of galaxies with flat or rising velocity dispersion profiles show higher dark matter fractions fDM both within 1 Re and 5 Re. Orbital anisotropy alone cannot account for the different observed trends of σ(R) and has a minor effect compared to variations of the mass profile. Galaxies with similar stellar mass M∗ that show different velocity dispersion profiles (from falling to rising) are successfully modelled with a variation of the halo mass Mh.

Flower cover and bee diversity estimate by drone RGB images

Worldwide, biodiversity is decreasing due to climate change, habitat fragmentation and agricultural intensification. Bees are essential crops pollinator, but their abundance and diversity are decreasing as well. For their conservation, it is necessary to assess the status of bee population. Field data collection methods are expensive and time consuming thus, recently, new methods based on remote sensing are used. In this study we tested the possibility of using flower cover diversity estimated by UAV images (FCD-UAV) to assess bee diversity and abundance in 10 agricultural meadows in the Netherlands. In order to do so, field data of flower and bee diversity and abundance were collected during a campaign in May 2021. Furthermore, RGB images of the areas have been collected using Unmanned Aerial Vehicle (UAV) and post-processed into orthomosaics. Lastly, Random Forest machine learning algorithm was applied to estimate FCD of the species detected in each field. Resulting FCD was expressed with Shannon and Simpson diversity indices, which were successively correlated to bee Shannon and Simpson diversity indices, abundance and species richness. The results showed a positive relationship between FCD-UAV and in-situ collected data about bee diversity, evaluated with Shannon index, abundance and species richness. The strongest relationship was found between FCD (Shannon Index) and bee abundance with R2=0.52. Following, good correlations were found with bee species richness (R2=0.39) and bee diversity (R2=0.37). R2 values of the relationship between FCD (Simpson Index) and bee abundance, species richness and diversity were slightly inferior (0.45, 0.37 and 0.35, respectively). Our results suggest that the proposed method based on the coupling of UAV imagery and machine learning for the assessment of flower species diversity could be developed into valuable tools for large-scale, standardized and cost-effective monitoring of flower cover and of the habitat quality for bees.

Il segnale EEG in un compito di ’center-out reaching’: analisi e decodifica a livello di scalpo e di corteccia

Nellâ analisi del segnale EEG, sono di particolare interesse i potenziali evento correlato (ERP), che descrivono la risposta cerebrale in seguito alla presentazione di stimoli o eventi esterni o interni. Questi potenziali non sono immediatamente visibili sul tracciato EEG grezzo, ed è necessario, oltre ad unâ��accurata fase di preprocessing, mediare (averaging) i segnali di molti trial ripetuti per visualizzare tali risposte nel tempo. Questo studio ha posto l' attenzione sugli ERP visuomotori generati in un compito di center-out reaching, che prevede il raggiungimento di uno tra cinque target, ognuno associato ad un LED, mediante il braccio dominante, con una tempistica scandita dalla presentazione di due stimoli visivi: lo stimolo preparatorio ¸ (che indica il target) e lo stimolo imperativo (che dà il via libera al movimento). I segnali ERP, ottenuti mediante la tecnica dellâ averaging, sono stati analizzati sia a livello di scalpo, considerando i segnali di elettrodo, sia a livello di corteccia, dopo risoluzione del problema inverso, e considerando rappresentazioni prima a livello di singoli dipoli corticali e quindi di intere regioni corticali (ROI). Inoltre, è stata applicato un metodo di deep learning (rete neurale convoluzionale) per decodificare il segnale EEG a livello di singolo trial, ovvero classificare il target coinvolto nello specifico trial. La decodifica è stata applicata sia ai segnali di scalpo sia ai segnali delle ROI corticali. Complessivamente i risultati mostrano ERP ben visibili a livello di scalpo e legati sia a processing visivo che motorio. Gli ERP a livello di ROI corticali sono più rumorosi e sembrano cogliere meno processing motorio rispetto al visivo, presumibilmente anche in conseguenza di alcune scelte metodologiche nella ricostruzione di segnali di ROI. In linea con questo, le performance di decodifica sono migliori a livello di scalpo che di ROI corticali.

Copy-paste data augmentation for domain transfer on traffic signs

City streets carry a lot of information that can be exploited to improve the quality of the services the citizens receive. For example, autonomous vehicles need to act accordingly to all the element that are nearby the vehicle itself, like pedestrians, traffic signs and other vehicles. It is also possible to use such information for smart city applications, for example to predict and analyze the traffic or pedestrian flows. Among all the objects that it is possible to find in a street, traffic signs are very important because of the information they carry. This information can in fact be exploited both for autonomous driving and for smart city applications. Deep learning and, more generally, machine learning models however need huge quantities to learn. Even though modern models are very good at gener- alizing, the more samples the model has, the better it can generalize between different samples. Creating these datasets organically, namely with real pictures, is a very tedious task because of the wide variety of signs available in the whole world and especially because of all the possible light, orientation conditions and con- ditions in general in which they can appear. In addition to that, it may not be easy to collect enough samples for all the possible traffic signs available, cause some of them may be very rare to find. Instead of collecting pictures manually, it is possible to exploit data aug- mentation techniques to create synthetic datasets containing the signs that are needed. Creating this data synthetically allows to control the distribution and the conditions of the signs in the datasets, improving the quality and quantity of training data that is going to be used. This thesis work is about using copy-paste data augmentation to create synthetic data for the traffic sign recognition task.

Traffic Signal Optimization of Urban Arterial Network by Vehicle-to-Infrastructure

Urbanization has occasionally been linked to negative consequences. Traffic light system in urban arterial networks plays an essential role to the operation of transport systems. The availability of new Intelligent Transportation System innovations paved the way for connecting vehicles and road infrastructure. GLOSA, or the Green Light Optimal Speed Advisory, is a recent integration of vehicle-to-everything (v2x) technology. This thesis emphasized GLOSA system's potential as a tool for addressing traffic signal optimization. GLOSA serves as an advisory to drivers, informing them of the speed they must maintain to reduce waiting time. The considered study area in this thesis is the Via Aurelio Saffi – Via Emilia Ponente corridor in the Metropolitan City of Bologna which has several signalized intersections. Several simulation runs were performed in SUMOPy software on each peak-hour period (morning and afternoon) using recent actual traffic count data. GLOSA devices were placed on a 300m GLOSA distance. Considering the morning peak-hour, GLOSA outperformed the actuated traffic signal control, which is the baseline scenario, in terms of average waiting time, average speed, average fuel consumption per vehicle and average CO2 emissions. A remarkable 97% reduction on both fuel consumption and CO2 emissions were obtained. The average speed of vehicles running through the simulation was increased as well by 7% and a time saved of 25%. Same results were obtained for the afternoon peak hour with a decrease of 98% on both fuel consumption and CO2 emissions, 20% decrease on average waiting time, and an increase of 2% in average speed. In addition to previously mentioned benefits of GLOSA, a 15% and 13% decrease in time loss were obtained during morning and afternoon peak-hour, respectively. Towards the goal of sustainability, GLOSA shows a promising result of significantly lowering fuel consumption and CO2 emissions per vehicle.