The adoption of upper-limb exoskeletons for rehabilitation offers a promising solution for the growing demand for effective therapeutic interventions, especially for people suffering from the motor consequences of stroke or other neuromuscular conditions. However, challenges such as high costs, limited clinical evidence of superiority over traditional therapy, and reduced therapist acceptance have hindered the widespread use of this technology. While the first two factors are largely due to the relative novelty of rehabilitation exoskeletons, and will likely improve with further time and technological advancement, overcoming therapists' lack of appreciation and trust towards robotics remains a challenge. This research seeks to overcome this barrier by developing a therapist-friendly exoskeleton that closely integrates the therapist's approach through innovative technological implementations. Specifically, it focuses on enabling the exoskeleton to mimic therapist movements, support and haptic- \\interaction strategies, and to provide real-time feedback about patient's conditions. Our therapist-friendly concept includes a Hidden Markov Models-based system to learn the most representative trajectory from the therapist's demonstrations, a novel force-mimicry framework to let the robot autonomously tune its support to the arm in a therapist-inspired way, and an augmented reality feedback platform to provide kinematic performance insights to the therapist. Experiments conducted with the AGREE and ANYexo robotic platforms to validate our frameworks reveal that these approaches improve therapist satisfaction, marking a significant step toward the broader clinical application of rehabilitation exoskeletons. The findings suggest that a synergistic design, combining therapist imitation and responsive feedback systems, can facilitate greater acceptance and efficacy of robotic rehabilitation solutions.
L'adozione di esoscheletri per la riabilitazione dell'arto superiore offre una soluzione promettente per rispondere alla crescente domanda di interventi terapeutici efficaci, in particolare per le persone che soffrono delle conseguenze motorie dell'ictus o di altre patologie neuromuscolari. Tuttavia, sfide come i costi elevati, la limitata evidenza clinica a dimostrazione della loro superiorità rispetto alla terapia tradizionale e la ridotta accettazione da parte dei terapisti hanno ostacolato la diffusione di questa tecnologia. Mentre i primi due fattori sono dovuti principalmente alla relativa "giovane età" degli esoscheletri riabilitativi e possiamo ipotizzare saranno risolti con il tempo e l'avanzamento tecnologico, superare la mancanza di apprezzamento e fiducia dei terapisti verso la robotica rimane una sfida. Questa ricerca mira a superare tale ostacolo sviluppando un esoscheletro "therapist-friendly", che integri strettamente l'approccio del terapista, implementando soluzioni tecnologiche che gli permettano di imitare i movimenti e le strategie di supporto e di interazione aptica del terapista, oltre a fornire feedback in tempo reale sulle condizioni del paziente. Il nostro concetto di "therapist-friendly" include: un sistema basato sul metodo di Machine Learning Hidden-Markov-Models (HMM), per apprendere la traiettoria più rappresentativa dalle dimostrazioni del terapista, un innovativo framework di imitazione della forza che permette al robot di adattare autonomamente il supporto fornito al braccio del paziente in modo ispirato all’approccio del terapista, e una piattaforma di feedback in realtà aumentata per fornire informazioni sulle prestazioni cinematiche del paziente. Gli esperimenti condotti con le piattaforme robotiche AGREE e ANYexo per validare i nostri framework rivelano che questi migliorano la soddisfazione dei terapisti, segnando un passo significativo verso una più ampia applicazione clinica degli esoscheletri riabilitativi. I risultati suggeriscono che un design sinergico, che combini l'imitazione del terapista e sistemi di feedback real-time, può facilitare una maggiore accettazione ed efficacia delle soluzioni robotiche per la riabilitazione.
Towards the adoption of upper limb rehabilitation exoskeletons in relevant environments
LUCIANI, BEATRICE
2024/2025
Abstract
The adoption of upper-limb exoskeletons for rehabilitation offers a promising solution for the growing demand for effective therapeutic interventions, especially for people suffering from the motor consequences of stroke or other neuromuscular conditions. However, challenges such as high costs, limited clinical evidence of superiority over traditional therapy, and reduced therapist acceptance have hindered the widespread use of this technology. While the first two factors are largely due to the relative novelty of rehabilitation exoskeletons, and will likely improve with further time and technological advancement, overcoming therapists' lack of appreciation and trust towards robotics remains a challenge. This research seeks to overcome this barrier by developing a therapist-friendly exoskeleton that closely integrates the therapist's approach through innovative technological implementations. Specifically, it focuses on enabling the exoskeleton to mimic therapist movements, support and haptic- \\interaction strategies, and to provide real-time feedback about patient's conditions. Our therapist-friendly concept includes a Hidden Markov Models-based system to learn the most representative trajectory from the therapist's demonstrations, a novel force-mimicry framework to let the robot autonomously tune its support to the arm in a therapist-inspired way, and an augmented reality feedback platform to provide kinematic performance insights to the therapist. Experiments conducted with the AGREE and ANYexo robotic platforms to validate our frameworks reveal that these approaches improve therapist satisfaction, marking a significant step toward the broader clinical application of rehabilitation exoskeletons. The findings suggest that a synergistic design, combining therapist imitation and responsive feedback systems, can facilitate greater acceptance and efficacy of robotic rehabilitation solutions.File | Dimensione | Formato | |
---|---|---|---|
PhD_thesis_Luciani_2025.pdf
non accessibile
Descrizione: PhD thesis Luciani: Towards the adoption of upper limb rehabilitation exoskeletons in relevant environments
Dimensione
75.76 MB
Formato
Adobe PDF
|
75.76 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/10589/233352