Damages to the central nervous system affect a growing number of people every year, particularly when coming to elderly people. The causes of these damages are different, stroke as first of them. Consequences for survivors include visuospatial and sensory-motor deficit, which limit the person from a physical and cognitive point of view, including collateral psychological effects. With reference to sensorimotor deficits at upper and lower limbs level, rehabilitation consists in training the injured limb, in increasing movements coordination and in learning new motor strategies, which exploit the great adaptive neuronal plasticity and the redundancy of nerve activations that characterize our brain. Duration and intensity of a task, as well as the repetitive, functional and stimulating characteristic of the performed motor task help making the therapy effective and guarantee visible outcomes in the functional recovery. From this point of view, the entry of robotics into the world of rehabilitation has represented a significant turning point. Indeed, it is possible to have a continuous re-adaptation of the therapy with the addition and application of new constraints that challenge patient’s progressively growing ability. Furthermore, the objective evaluation of the effects, of therapy progresses and of any pharmacological treatments are made possible by the automated collection of data and their corresponding analysis. Thus it is possible to ensure a more accurate short and long term diagnosis. In this way, the patient him/herself can also become aware of his/her progresses. Functional improvements self-awareness is important because the patient, as a disabled person, needs to be motivated and encouraged to continue the therapy and to carry it out longer and more intensively. Nowadays, the development of robotic devices dedicated to rehabilitation is on the rise, but their production and use has still very high costs to be considered accessible to a large number of people. Therefore, as it happens in any other technological field, through the time, the continuous development and evolution of knowledge will make this type of rehabilitation increasingly accessible and low-cost, considering which is generally the population target that will have to use it. The control strategies that a robotic devices adopt to provoke neuronal plasticity are of different types. In particular, assistive controllers are based on a rehabilitation paradigm of "active assistance" in which the robot mimics the assistance provided by the physiotherapist: it is therefore an external, physical assistance, which helps the patient to perform and complete the task required by the exercise. The subcategories of assistance are classified according to the sensor with which they are equipped. The assistance can be impedance-based, triggered, EMG-based or performance-based with the adaptation of task parameters. In particular the EMG-based control methods are one of the most recurrent techniques used in biorobotics applications, and they are very effective in measuring the intensity of human motion directly and in reflecting the muscle activity of the user. it is necessary to underline that there are however some challenges associated with this techniques depending upon the user application. In fact, the most considerable problem of EMG-based methods is that they can not be used in case of the user is unable to generate sufficient muscle signals, as people with totally paralyzed limb, who in fact are not the target patients of this project. The aim of this work is to design and implement an EMG-based control for a 5-degrees of freedom upper-limb powered exoskeleton, which has been designed to the therapy of subjects affected by neuromotor diseases.
I danni al sistema nervoso centrale colpiscono un numero crescente di persone ogni anno, soprattutto in età avanzata; le cause di questi danni possono essere molteplici, fra tutte l'infarto. Le conseguenze per i sopravvissuti a tale trauma comprendono deficit visuospaziale e sensoriale-motorio, che limita la persona dal punto di vista fisico, compromissioni cognitive e effetti psicologici collaterali. I principali deficit sensoriali-motori riguardano le funzionalità degli arti superiori e inferiori. La riabilitazione in questi casi consiste nell'allenamento dell'arto ferito, nell'aumentare il coordinamento dei movimenti e nell'apprendimento di nuove strategie motorie, che sfruttano la grande plasticità neuronale adattiva e la ridondanza delle attivazioni nervose, che caratterizzano il nostro cervello. Per i pazienti la durata e l'intensità dell'esecuzione di un compito, nonché le caratteristiche ripetitive, funzionali e stimolanti, aiutano a rendere efficace la terapia e garantiscono risultati visibili nel recupero funzionale. Da questo punto di vista, l'ingresso della robotica nel mondo della riabilitazione ha rappresentato una svolta significativa. Grazie ad essa ora è possibile avere un continuo riadattamento della terapia, con l'aggiunta e l'applicazione di nuovi vincoli che sfidano la capacità di recupero progressiva del paziente. Inoltre la valutazione obiettiva degli effetti, dei progressi della terapia e di eventuali trattamenti farmacologici è resa possibile dalla raccolta automatizzata di dati e dalla loro analisi corrispondente, garantendo una diagnosi più accurata a breve e lungo termine. In questo modo il paziente stesso può anche prendere coscienza dei suoi progressi e questo è importante perché il paziente, in quanto persona disabile, deve essere motivato e incoraggiato sia a continuare la terapia, sia a portarla avanti il più a lungo possibile e in modo più intenso. Oggi siamo in una fase di sviluppo di dispositivi robotici dedicati alla riabilitazione, ma la loro produzione e il loro utilizzo hanno ancora costi decisamente elevati perchè siano considerati dispositivi accessibili a un gran numero di persone. Pertanto, come accade in qualsiasi altro campo tecnologico, con il passare del tempo lo sviluppo e l'evoluzione della conoscenza renderanno questo tipo di riabilitazione sempre più accessibile e a basso costo, tenendo sempre presente il target di utilizzatori finali. Le strategie di controllo che i dispositivi robotici adottano per stimolare la plasticità neuronale sono di diversi tipi. In particolare, i controller assistivi si basano su un paradigma riabilitativo di "assistenza attiva" in cui il robot imita l'assistenza fornita dal fisioterapista: si tratta quindi di un'assistenza fisica esterna, che aiuta il paziente a svolgere e completare il compito richiesto dal esercizio. Le sottocategorie di assistenza sono classificate in base al sensore di cui sono dotate. L'assistenza può essere impedance-based, triggered, EMG-based or performance-based con l'adattamento dei parametri dell'attività. In particolare, i metodi di controllo basati su EMG sono una delle tecniche più ricorrenti utilizzate nelle applicazioni biorobotiche e sono molto efficaci nel misurare direttamente l'intensità del movimento umano e nel riflettere l'attività muscolare dell'utente. Tuttavia, è necessario sottolineare che ci sono alcune sfide associate a queste tecniche a seconda dell'applicazione dell'utente. Infatti, il problema più rilevante dei metodi basati su EMG è che non possono essere utilizzati nel caso in cui l'utente non sia in grado di generare segnali muscolari sufficienti, cioè persone con arto totalmente paralizzato, che però non sono i pazienti target di questo progetto. Lo scopo di questo lavoro di tesi è quello di progettare e implementare un controllo basato su EMG per un esoscheletro motorizzato degli arti superiori a 5 DOF, che è stato progettato per la terapia di soggetti affetti da malattie neuromotorie.
Preliminary design of an EMG control for an upper limb rehabilitation exoskeleton
GUARNIERI, GIOVANNI
2018/2019
Abstract
Damages to the central nervous system affect a growing number of people every year, particularly when coming to elderly people. The causes of these damages are different, stroke as first of them. Consequences for survivors include visuospatial and sensory-motor deficit, which limit the person from a physical and cognitive point of view, including collateral psychological effects. With reference to sensorimotor deficits at upper and lower limbs level, rehabilitation consists in training the injured limb, in increasing movements coordination and in learning new motor strategies, which exploit the great adaptive neuronal plasticity and the redundancy of nerve activations that characterize our brain. Duration and intensity of a task, as well as the repetitive, functional and stimulating characteristic of the performed motor task help making the therapy effective and guarantee visible outcomes in the functional recovery. From this point of view, the entry of robotics into the world of rehabilitation has represented a significant turning point. Indeed, it is possible to have a continuous re-adaptation of the therapy with the addition and application of new constraints that challenge patient’s progressively growing ability. Furthermore, the objective evaluation of the effects, of therapy progresses and of any pharmacological treatments are made possible by the automated collection of data and their corresponding analysis. Thus it is possible to ensure a more accurate short and long term diagnosis. In this way, the patient him/herself can also become aware of his/her progresses. Functional improvements self-awareness is important because the patient, as a disabled person, needs to be motivated and encouraged to continue the therapy and to carry it out longer and more intensively. Nowadays, the development of robotic devices dedicated to rehabilitation is on the rise, but their production and use has still very high costs to be considered accessible to a large number of people. Therefore, as it happens in any other technological field, through the time, the continuous development and evolution of knowledge will make this type of rehabilitation increasingly accessible and low-cost, considering which is generally the population target that will have to use it. The control strategies that a robotic devices adopt to provoke neuronal plasticity are of different types. In particular, assistive controllers are based on a rehabilitation paradigm of "active assistance" in which the robot mimics the assistance provided by the physiotherapist: it is therefore an external, physical assistance, which helps the patient to perform and complete the task required by the exercise. The subcategories of assistance are classified according to the sensor with which they are equipped. The assistance can be impedance-based, triggered, EMG-based or performance-based with the adaptation of task parameters. In particular the EMG-based control methods are one of the most recurrent techniques used in biorobotics applications, and they are very effective in measuring the intensity of human motion directly and in reflecting the muscle activity of the user. it is necessary to underline that there are however some challenges associated with this techniques depending upon the user application. In fact, the most considerable problem of EMG-based methods is that they can not be used in case of the user is unable to generate sufficient muscle signals, as people with totally paralyzed limb, who in fact are not the target patients of this project. The aim of this work is to design and implement an EMG-based control for a 5-degrees of freedom upper-limb powered exoskeleton, which has been designed to the therapy of subjects affected by neuromotor diseases.File | Dimensione | Formato | |
---|---|---|---|
TESI_GIOVANNI_GUARNIERI_895984_definitiva.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
64.62 MB
Formato
Adobe PDF
|
64.62 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/165481