Foot transimitted vibration (FTV) refers to the mechanical stimulus transmitted to the human body as a whole through the supporting surface. FTV has been recently classified as a separate cathegory from Whole-Body vibration (WBV), that is the vibration transmitted to the subject through the seat. FTV and WBV have both been associated with health risks, including fatigue, discomfort, and an increased risk of injury or aggravation of existing conditions such as herniated discs or cardiovascular disease. This work focused on the development of a walking humanoid model to predict the effects of FTV on balance and fall probability. The model benefits from a realistic real-time physical simulation and a direct dynamics approach and is derived from 3D models available in the literature to generate a human-like gait. It has 23 degrees of freedom, which are controlled by proportional and derivative controllers and muscle-tendon units (MTUs). There are 8 MTUs to control sagittal motion for each leg. The control parameters of the PD controllers and the MTUs are optimised thanks to a genetic algorithm called Covariance Matrix Adaptation Evolutionary Strategy (CMAES). The controller has been optimised in the case of sinusoidal roll WBV with 4° amplitude and 0.25-1 Hz frequencies. In the second part of the work, 40 healthy subjects were exposed to 71 different vibration conditions, including sinusoidal and random stimuli with roll, pitch and vertical motion. Participants were exposed to vibration frequency between 0.25 and 1 Hz; the vibration amplitude was between 4° and 8° for roll and pitch, and 4 cm and 8 cm for vertical motion. The main result is that sinusoidal roll motion at 1 Hz and 8° significantly increased step width and step frequency and decreased step stance and step length in up to 40% of subjects. The experimental gait metrics during roll FTV were compared with the corresponding numerical simulations obtained with the model developed in the first part, showing similar adaptation patterns to vibration. Finally, we investigated the effect of FTV on cognitive performance with two sets of experiments. In the first, we studied the effect of vibration along different axes on the response time (RT) of 25 standing subjects during a customised psychomotor vigilance task (PVT). A second series of experiments focused on analysing the effects of medio-lateral vibration on the RT of 20 walking subjects. The 10% increase in mean RT during walking exposed to WBV, compared to baseline, suggests that the cognitive reaction is delayed due to: additional musculoskeletal recruitment, head motion due to the presence of WBV and increased attention to peripheral cues to maintain the balance. However, walking with WBV doesn’t necessarily reduce vigilance compared to walking without WBV or standing with WBV.
La vibrazione transimmessa dal piede (FTV) si riferisce allo stimolo meccanico trasmesso al corpo umano nel suo complesso attraverso la superficie di appoggio. La FTV è stata recentemente classificata come una categoria separata dalle vibrazioni trasmesse al corpo intero (WBV), ovvero le vibrazioni trasmesse al soggetto attraverso il sedile. Sia la FTV che la WBV sono state associate a rischi per la salute, tra cui affaticamento, disagio e aumento del rischio di lesioni o di aggravamento di patologie già esistenti, come l'ernia del disco o le malattie cardiovascolari. Questo lavoro si è concentrato sullo sviluppo di un modello umanoide di camminata per prevedere gli effetti del FTV sull'equilibrio e sulla probabilità di caduta. Il modello beneficia di una simulazione fisica realistica in tempo reale e di un approccio dinamico diretto: esso è derivato da modelli 3D disponibili in letteratura per generare un'andatura simile a quella umana. Dispone di 23 gradi di libertà, controllati da regolatori proporzionali e derivati e da unità muscolo-tendinee (MTU). Le unità muscolo-tendinee (MTU) sono 8 e controllano il movimento sagittale di ciascuna gamba. I parametri di controllo dei regolatori PD e delle unità muscolo-tendinee sono ottimizzati grazie a un algoritmo genetico chiamato Covariance Matrix Adaptation Evolutionary Strategy (CMAES). Il controllore è stato ottimizzato nel caso di WBV a rollio sinusoidale con ampiezza di 4° e frequenze di 0,25-1 Hz. Nella seconda parte del lavoro, 40 soggetti sani sono stati esposti a 71 diverse condizioni di vibrazione, tra cui stimoli sinusoidali e casuali con rollio, beccheggio e movimento verticale. I partecipanti sono stati esposti a una frequenza di vibrazione compresa tra 0,25 e 1 Hz; l'ampiezza della vibrazione era compresa tra 4° e 8° per il rollio e il beccheggio e tra 4 cm e 8 cm per il movimento verticale. Il risultato principale è che il movimento sinusoidale di rollio a 1 Hz e 8° ha aumentato significativamente l'ampiezza del passo e la frequenza del passo e ha diminuito l'ampiezza e la lunghezza del passo fino al 40% dei soggetti. Le metriche sperimentali dell'andatura durante il roll FTV sono state confrontate con le corrispondenti simulazioni numeriche ottenute con il modello sviluppato nella prima parte, mostrando modelli di adattamento alle vibrazioni simili. Infine, abbiamo studiato l'effetto del FTV sulle prestazioni cognitive con due serie di esperimenti. Nel primo, abbiamo studiato l'effetto delle vibrazioni lungo diversi assi sul tempo di risposta (RT) di 25 soggetti in piedi durante un compito di vigilanza psicomotoria (PVT) personalizzato. Una seconda serie di esperimenti si è concentrata sull'analisi degli effetti delle vibrazioni medio-laterali sul RT di 20 soggetti che camminavano. L'aumento del 10% del RT medio durante la camminata esposta a WBV, rispetto al basale, suggerisce che la reazione cognitiva è ritardata a causa di: reclutamento muscolo-scheletrico aggiuntivo, movimento della testa dovuto alla presenza di WBV e maggiore attenzione alle indicazioni periferiche per mantenere l'equilibrio. Tuttavia, la camminata con WBV non riduce necessariamente la vigilanza rispetto a quella che si ha in presenza di WBV.
Modelling and experiments on the effects of foot-transmitted vibration while walking
Marelli, Stefano
2023/2024
Abstract
Foot transimitted vibration (FTV) refers to the mechanical stimulus transmitted to the human body as a whole through the supporting surface. FTV has been recently classified as a separate cathegory from Whole-Body vibration (WBV), that is the vibration transmitted to the subject through the seat. FTV and WBV have both been associated with health risks, including fatigue, discomfort, and an increased risk of injury or aggravation of existing conditions such as herniated discs or cardiovascular disease. This work focused on the development of a walking humanoid model to predict the effects of FTV on balance and fall probability. The model benefits from a realistic real-time physical simulation and a direct dynamics approach and is derived from 3D models available in the literature to generate a human-like gait. It has 23 degrees of freedom, which are controlled by proportional and derivative controllers and muscle-tendon units (MTUs). There are 8 MTUs to control sagittal motion for each leg. The control parameters of the PD controllers and the MTUs are optimised thanks to a genetic algorithm called Covariance Matrix Adaptation Evolutionary Strategy (CMAES). The controller has been optimised in the case of sinusoidal roll WBV with 4° amplitude and 0.25-1 Hz frequencies. In the second part of the work, 40 healthy subjects were exposed to 71 different vibration conditions, including sinusoidal and random stimuli with roll, pitch and vertical motion. Participants were exposed to vibration frequency between 0.25 and 1 Hz; the vibration amplitude was between 4° and 8° for roll and pitch, and 4 cm and 8 cm for vertical motion. The main result is that sinusoidal roll motion at 1 Hz and 8° significantly increased step width and step frequency and decreased step stance and step length in up to 40% of subjects. The experimental gait metrics during roll FTV were compared with the corresponding numerical simulations obtained with the model developed in the first part, showing similar adaptation patterns to vibration. Finally, we investigated the effect of FTV on cognitive performance with two sets of experiments. In the first, we studied the effect of vibration along different axes on the response time (RT) of 25 standing subjects during a customised psychomotor vigilance task (PVT). A second series of experiments focused on analysing the effects of medio-lateral vibration on the RT of 20 walking subjects. The 10% increase in mean RT during walking exposed to WBV, compared to baseline, suggests that the cognitive reaction is delayed due to: additional musculoskeletal recruitment, head motion due to the presence of WBV and increased attention to peripheral cues to maintain the balance. However, walking with WBV doesn’t necessarily reduce vigilance compared to walking without WBV or standing with WBV.File | Dimensione | Formato | |
---|---|---|---|
PhD_Thesis_Marelli_2023.pdf
non accessibile
Dimensione
19.4 MB
Formato
Adobe PDF
|
19.4 MB | Adobe PDF | Visualizza/Apri |
Subject_Reports.zip
non accessibile
Dimensione
77.75 MB
Formato
Unknown
|
77.75 MB | Unknown | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/213792