67 research outputs found
AI-Based Techniques for Enhancing Human-Robot Collaboration
No full textQuesta tesi di dottorato promuove la collaborazione tra umani e robot sviluppando un assistente robotico adattivo capace di garantire un’interazione sicura, fluida e produttiva. Nell’industria contemporanea ci sono processi dove non è possibile
avere un’automazione totale, gli operatori apportano competenze e conoscenze indispensabili a processi specializzati, mentre i robot possono ridurre lo sforzo fisico,
migliorare l’efficienza del flusso di lavoro e contribuire al benessere dell’operatore.
Questa ricerca utilizza tecniche di intelligenza artificiale, tra cui machine learning,
deep learning e large multimodal models, per creare postazioni di lavoro collaborative in cui umani e robot lavorano fianco a fianco in modo intuitivo, condividendo
compiti e spazi operativi.
I sistemi robotici sviluppati sono dotati di capacità di percezione, che consentono
loro di rilevare, interpretare e rispondere agli input e umani in tempo reale. Grazie
all’uso di dati provenienti da sensori e telecamere, il robot si adatta dinamicamente
ai movimenti e alle intenzioni dell’operatore, dando priorità alla sicurezza e alla
fluidità dell’interazione, ottimizzando allo stesso tempo il flusso di lavoro e riducendo
lo sforzo umano. Queste capacità sono state testate in diverse postazioni robotiche,
come in operazioni di assemblaggio e produzione, dimostrando come il robot possa
migliorare la sicurezza e l’efficienza.
Un aspetto chiave di questo lavoro è stato lo sviluppo di un sistema di percezione
della postura del corpo umano mediante telecamere RGB-D, capace di monitorare
in tempo reale la posizione delle articolazioni a una frequenza di 18 Hz, con una
precisione verificata attraverso un sistema basato su IMU. Identificando la posizione
dell’operatore, il robot può mantenere una distanza di sicurezza e adattarsi dinamicamente per evitare collisioni in diversi scenari operativi. Basandosi su questo sistema di tracciamento e di evitamento degli ostacoli, la tesi integra ulteriori funzionalità come la selezione di strumenti tramite gesti e il recupero automatico del tool
usato dall’operatore, consentendo al robot di rispondere ai comandi dell’operatore in modo fluido e di prevenire collisioni, ribaltando così il tradizionale modello di
interazione uomo-macchina.
In collaborazione con l’Università di Coimbra, la ricerca ha inoltre sviluppato il
sistema D-RMGPT, che utilizza large multimodal models per migliorare le attività
di assemblaggio. Questo sistema permette di ridurre i tempi di assemblaggio del 33%
per operatori principianti e offre una grande flessibilità grazie a prompt basati su
immagini, eliminando la necessità di un ampio addestramento dei dati. Inoltre, un
caso di studio che riguarda l’esecuzione di un processo, in questo caso la preparazione
di una pizza, dimostra l’uso del modello GPT per comporre la sequenza del processo
e gestire le azioni da far compiere al robot. Infatti, ricevendo gli input dell’operatore,
come la selezione degli ingredienti tramite voce/gesti, e riconoscendo le azioni umane,
il robot può decidere le sue azioni di risposta di conseguenza per fornire assistenza
all’operatore durante il processo. Complessivamente, questi contributi stabiliscono
un quadro robusto per una collaborazione uomo-robot adattabile, in tempo reale, e
intuitiva, che migliora così il benessere dell’operatore e la produttività.This PhD thesis advances human-robot collaboration by developing an adaptive
robotic assistant that ensures safe, seamless, and productive collaboration between
humans and machines. In modern industry is not possible to apply full automation in some contests, humans contribute with essential knowledge and skills to
specialized processes, while robots can assist by reducing physical strain, improving workflow, and enhancing operator well-being. This research leverages artificial
intelligence techniques such as machine learning, deep learning, and large multimodal models to create a collaborative workstations where humans and robots work
together intuitively, sharing tasks and workspace as one team.
The proposed robotics systems are equipped with advanced perception capabilities, allowing them to detect, interpret, and respond to human actions and inputs in
real time. Using sensors and cameras data, the robot adapts dynamically to the operator’s movements and intentions, prioritizing safety and fluency while optimizing
workflow and minimizing the human effort. This capability was tested in various
real-world robotic workstations, such as assembly and manufacturing tasks, where
the robot provided safety and efficiency enhancements.
A starting point of this work includes a human body pose perception system using RGB-D cameras, tracking joint positions in real time at 18 Hz with accuracy
validated against an IMU wearable based system. By identifying the operator’s
position, the robot maintains a safe distance and dynamically avoids collisions in
various scenarios. Building on this skeleton tracking and obstacle avoidance framework, the thesis further integrates gesture-based tool selection and automatic tool
retrieval, allowing the robot to respond fluidly to operator commands and to avoid
it, effectively reversing the traditional human-machine interaction model.
The research further developed, in collaboration with the University of Coimbra,
D-RMGPT, a system employing large multimodal models for assembly tasks. This
system reduces assembly time by 33% for novices and offers high flexibility through
viiimage-based prompts without the need for extensive data training. Additionally, a
case study involving a process execution, in this circumstance as an example a pizza
preparation, demonstrates the use of a GPT model to compose the process sequence
and manage the robot actions. In fact, receiving the operator’s inputs such as
ingredient selection via voice/gestures and recognizing the human actions, the robot
can adjust its responses accordingly to provide assistance to the operator during
the process. Together, all these contributions establish a robust framework for
adaptable, intuitive real-time human-robot collaboration that improves the human
wellbeing and productivit
Long-term follow-up of retropupillary iris-claw intraocular lens implantation: A retrospective analysis Cataract and refractive surgery
No full textBackground: The ideal intraocular lens in cases of aphakia without capsular support is debated. Choices include anterior chamber lenses, iris- or scleral-sutured lenses, and iris-claw lenses. Our aim was to report our long-term evaluation of the use of retropupillary implantation of the Artisan iris-claw intraocular lens (RPICIOL) in several aphakic conditions without capsular support. Methods: A retrospective analysis of consecutive 320 eyes of 320 patients (222 males and 98 females) without capsular support in which we performed RPICIOL implantation in post-traumatic aphakia (141 eyes, group 1), post-cataract surgery aphakia (122 eyes, group 2), and in cases in which penetrating keratoplasty was associated with vitrectomy (57 eyes, group 3). Either anterior or posterior vitrectomy procedures were performed with 20-, 23-, or 25-gauge techniques for different associated anterior or posterior segment indications. We reviewed the refractive outcome, anatomical outcome, long-term stability of the implants, and possible long-term complications. Results: The mean patient age was 59.7 years (range, 16-84 years) in group 1; 60.1 years (range, 14-76 years) in group 2; and 65.8 years (range, 25-71.5 years) in group 3. The mean follow-up time was 5.3 years (range, 1 month to 8 years). At the end of the follow-up period, the mean post-operative best-corrected LogMAR visual acuity was 0.6 (range, perception of light to 0.3) in group 1; 0.3 (range, 0.5-0.1) in group 2; and 0.6 (range, hand movement to 0.2) in group 3. Disenclavation of RPICIOLs occurred in three cases because of slippage of one of the iris-claw haptics and spontaneous complete posterior dislocation occurred in one case. One case presented with retinal detachment, and no cases of uveitis were observed. Eight cases complained of chronic dull pain, and severe iridodonesis was seen in five cases. One case of post-operative macular edema was observed without post-operative increase in the mean intraocular pressure. There was no statistically different change in the endothelial cell density (cells/mm2) at the end of the follow-up period. Conclusions: RPICIOL for secondary implantations is a valid alternative strategy to scleral-fixated or angle-supported IOL implantation
Akreos Adapt AO Intraocular Lens Opacification: A Case Report.
No full textA 79-year-old hyperglycemic patient was referred to the 'Santa Maria delle Croci' Hospital, Ravenna, Italy. He presented with visual impairment in the right eye. Four years ago, he had had an uneventful cataract surgery in the right eye. We observed an opacification of the intraocular lens (IOL) causing significant visual disturbance. The IOL was exchanged. Unfortunately, pathologic analysis was not performed. Patient-related factors such as hyperglycemia and hypertension might have been responsible for the opacification. To our knowledge, there is only one previous report of opacification of the Akreos Adapt AO IOL
Performing the City: Space, Movement, and Memory in O Ben'Groes at Droed Amser
No full textIn considering some of the aspects that are brought to the site-specific process, this article explores how theatre renegotiates patterns of intraurban movement, enacting complex approaches to space and memory. The focus is on O Ben'Groes at Droed Amser, a Theatr Genedlaethol Cymru and National Theatre Wales production, created in collaboration with BBC Cymru Wales and BBC Arts in 2020 as part of Theatr Genedlaethol Cymru's new programme in response to the COVID-19 emergency and the challenge of creating original dramatic work during lockdown. By allowing virtual audiences to join author and poet Karen Owen on a bus journey from her home and the street where she grew up to Bangor city centre, the production articulated the experience of the city in terms of individual and collective memory, bringing together issues of performance, representation, history, and heritage to reveal alternative layers to the reality of the urban landscape. Memory, I argue, emerged from this production as a performative construct open to renegotiation through a range of present relationships to landscape. In addition to this, the production also offered an alternative to the privileged figure of the walker, performing a subversion of codified patterns of movement
Clinical risk management in eye outpatient surgery: a new surgical safety checklist for cataract surgery and intravitreal anti-VEGF injection
No full textAbstract
BACKGROUND:
A database study to test a new model of surgical safety checklist for eye surgery in the outpatient operating room, especially for cataract surgery and intravitreal anti-VEGF injections.
METHODS:
WHO Surgical Safety Checklist analysis, and changes to obtain a customized surgical safety checklist for eye surgery. Testing of the new checklist in the outpatient operating room of the Institute of Ophthalmology, University of Modena during a period of 4 months (January-April 2011). All cataract surgery and intravitreal anti-VEGF operations were included in the study, and controlled by the new surgical safety checklist. The percentage of answers to each safety check was calculated to obtain an estimate of adherence to our new checklist.
RESULT:
Eight hundred and forty nine procedures (390 cataract, 452 anti-VEGF injections, seven combined surgery) were analyzed. The study showed a high level of adherence to the majority of the safety checks in the checklist. Important differences were identified in consent confirmation (correct confirmation in 99.76 %, no confirmation in 0.24 %), surgical site marking (correct marking in 99.29 %, no marking in 0.71 %), patient cooperation during operation (95.17 % of patients were cooperative, 4.83 % of patients were uncooperative), adherence to antibiotic prophylaxis (correct prophylaxis in 93.17 %, no prophylaxis in 6.83 %), critical situation during surgical procedures (no critical situation in 91.76 %, critical situation in 8.24 %)
CONCLUSION:
The study showed a high level of adherence to the checklist. A surgical safety checklist could improve the management of an eye surgery operating room
Advancing human–robot collaboration in handcrafted manufacturing: cobot-assisted polishing design boosted by virtual reality and human-in-the-loop
No full textIndustry 5.0 envisions a future where seamless collaboration between humans and robots enhances efficiency, innovation, and coevolution. While collaborative robots have found widespread applications in manufacturing, particularly in tasks like pick-and-place and assembly, their integration into handcrafted manufacturing processes presents unique challenges. This article focuses on advancing technology in the less-explored field of cobot-assisted handcrafted manufacturing, specifically in the fashion industry, with a priority on reducing work-related risks. In handcrafted processes, which often involve intricate and artistic work, cobots face challenges related to nuanced decision-making, adaptability to customizations, and the need for precise manual dexterity. The study delves into the cobot-assisted polishing of leather shoes, addressing issues associated with product delicacy, process and knowledge formalization, versatility, and integration into existing manufacturing processes. To overcome these challenges, the research proposes the application of cobots in the initial polishing phase, which is the most physically demanding, allowing artisans to focus on finalization, quality control, and process supervision. The study also applies the concept of human-in-the-loop (HITL) and virtual reality simulation to optimize collaboration, ensuring safety, ergonomics, and efficiency. The article contributes to the scientific and industrial communities by pioneering the study of collaborative robotics in craftsmanship, successfully implementing human–robot collaboration (HRC) in an industrial setting, demonstrating the effectiveness of virtual simulation and HITL, and prioritizing human factors throughout the design and development of HRC. The insights gained from this research are crucial for achieving practical solutions in industrial environments while aligning with the performance objectives of companies and workers’ well-being
MODIFICATION of ADJUSTABLE MACULAR BUCKLING with 29-G CHANDELIER LIGHT for OPTIMAL POSITIONING in HIGHLY MYOPIC EYES with MACULAR HOLE
No full textPurpose: To evaluate the efficacy of the modification of Adjustable Macular Buckling device in the treatment of myopic macular hole retinal detachment with posterior staphyloma. Methods: Four consecutive patients suffering from myopic macular detachment with macular hole were treated using the macular buckling procedure. An Adjustable Macular Buckling device was used in all four cases and was modified using a 29-gauge optical fiber to illuminate its macular plate. Results: Optical coherence tomography showed successful retinal reattachment and closure of the macular hole after the buckling procedure. The macular plate of the buckling device was properly positioned in all four patients. No complications were observed. Conclusion: The modification of the macular buckling device improves the accuracy of its positioning by illuminating its macular plate
Simultaneous correction of post-traumatic aphakia and aniridia with the use of artificial iris and IOL implantation
No full textAbstract
BACKGROUND:
Combined post-traumatic aniridia and aphakia demand extensive and complex reconstructive surgery. We present our approach for simultaneous correction of this surgical situation with the use of the ArtificialIris (Dr. Schmidt Intraocularlinsen GmbH, Germany) with a foldable acrylic IOL Lentis L-313 (Oculentis, GmbH, Germany) sutured to its surface. The novelty (our first operation was on June 2010) of this surgical technique is based on the combined use of foldable (with closed haptics) IOL and Artificialiris to correct post-traumatic aniridia and aphakia.
METHODS:
Four consecutive cases of combined post-traumatic lesions of iris and lens, corrected with complex device ArtificialIris and foldable IOL. In two cases, the compound implant was sutured to the sclera in sulcus during the penetrating keratoplasty; in another case, it was positioned through a corneal incision of about 5.0 mm with transscleral fixation, and in one patient with preserved capsular support and possibility of IOL in-the-bag implantation the ArtificialIris was placed in sulcus sutureless through a clear corneal tunnel.
RESULTS:
Maximal follow-up was 6 months. The complex device was placed firmly fixed within the sulcus, including in the eye implanted without sutures, and showed a stable and centered position without any tilt or torque.
CONCLUSION:
Management of post-traumatic aniridia combined with aphakia by haptic fixation of a foldable acrylic IOL on a foldable iris prosthesis appears to be a promising approach which gives the surgeon the possibility to correct a complex lesion with one procedure, which is less traumatic and faster. Existence of foldable materials, both iris and IOL, permits relatively small corneal incisions (4.0-5.0 mm). Moreover, the custom-tailored iris prosthesis gives a perfect aesthetic result
Disseminating Collaborative Robotics and Artificial Intelligence Through a Board Game Demo
No full text- …
