Human robot collaboration is an emerging field in the industrial context, since it combines the potentiality and skills of humans and the precision/accuracy of robots in order to improve performance, higher production efficiency and reduce products life cycles. The co-existence of human and robot in the same workplace has however to be ensured by a stable and safe interaction between the two, and at the same time has to guarantee that performance of collaboration are not lowered: regarding the latter, fatigue and ergonomics of the operator can be inspected with the aim to reduce human tiredness during tasks execution. A simple fatigue model based on the electromyography signal coming from the affected muscles is proposed and analysed in this work. To ensure stability of the system, the admittance control method is exploited with a proper selection of its gains: since the robot interacts with an external environment, represented by the operator, also the human counterpart has to be taken into account. The proposed control method designs a novel methodology for adapting on-line the gains of an admittance controlled robot, based on the current human arm impedance. Firstly, the human arm is characterized through experimental identi cation of its impedance parameters; secondly the adaptive admittance control algorithm is derived to tune its gains so that the passivity of human robot interaction system is guaranteed and the stability preserved. Moreover, the effects of robot links and joints elasticity on the algorithm are evaluated. Finally, the proposed approach is validated in a simulation environment testing some simple lead-through programming tasks. In addition, the fatigue model is exploited to check the operator muscles tiredness.
La collaborazione tra uomo e robot è un campo emergente nel contesto industriale, poiché permette di combinare le abilità umane e la precisione ed accuratezza del robot per raggiungere alti livelli di performance, ed allo stesso tempo, di far crescere l'efficienza riducendo i cicli di produzione. La coesistenza tra uomo e robot nel medesimo spazio lavorativo deve essere però assicurata da un'interazione stabile e sicura, e deve contemporaneamente garantire che l'efficienza della collaborazione non venga compromessa: per questo motivo la stanchezza dell'operatore, in termini di fatica muscolare, può essere tenuta sotto controllo allo scopo di essere minimizzata. In questa tesi è quindi proposto ed analizzato un semplice modello che stimi la fatica dell'operatore basandosi sul segnale elettromiogra fico (EMG) proveniente dai muscoli interessati. Per assicurare la stabilità del sistema uomo robot viene utilizzato il controllo di ammettenza,dove la selezione dei guadagni è cruciale in quanto il robot interagisce con l'ambiente esterno, in questo caso rappresentato all'operatore umano. Il metodo di controllo proposto tiene conto di ciò, e viene quindi progettato un nuovo regolatore che adatti simultaneamente i guadagni del fi ltro, basandosi sull'impedenza del braccio umano. Per prima cosa, attraverso una serie di esperimenti, vengono identi ficati i parametri di impedenza del braccio; in secondo luogo l'algoritmo adattivo, necessario a regolare il valore dei guadagni, viene derivato in modo che la passività del sistema e la sua stabilità siano garantite. In aggiunta, viene valutato l'effetto dell'elasticità di giunti e collegamenti del robot sull'algoritmo. In fine, il metodo proposto è convalidato in un ambiente di simulazione, dove vengono eseguite semplici operazioni di manual guidance. Inoltre, viene testato il modello di fatica muscolare, precedentemente analizzato, per controllare il livello di stanchezza dell'operatore.
Design of an adaptive admittance controller in manual guidance collaborative operations
BETTINELLI, FEDERICO
2018/2019
Abstract
Human robot collaboration is an emerging field in the industrial context, since it combines the potentiality and skills of humans and the precision/accuracy of robots in order to improve performance, higher production efficiency and reduce products life cycles. The co-existence of human and robot in the same workplace has however to be ensured by a stable and safe interaction between the two, and at the same time has to guarantee that performance of collaboration are not lowered: regarding the latter, fatigue and ergonomics of the operator can be inspected with the aim to reduce human tiredness during tasks execution. A simple fatigue model based on the electromyography signal coming from the affected muscles is proposed and analysed in this work. To ensure stability of the system, the admittance control method is exploited with a proper selection of its gains: since the robot interacts with an external environment, represented by the operator, also the human counterpart has to be taken into account. The proposed control method designs a novel methodology for adapting on-line the gains of an admittance controlled robot, based on the current human arm impedance. Firstly, the human arm is characterized through experimental identi cation of its impedance parameters; secondly the adaptive admittance control algorithm is derived to tune its gains so that the passivity of human robot interaction system is guaranteed and the stability preserved. Moreover, the effects of robot links and joints elasticity on the algorithm are evaluated. Finally, the proposed approach is validated in a simulation environment testing some simple lead-through programming tasks. In addition, the fatigue model is exploited to check the operator muscles tiredness.File | Dimensione | Formato | |
---|---|---|---|
2020_04_Bettinelli.pdf
accessibile in internet per tutti
Descrizione: Testo completo della tesi
Dimensione
2.69 MB
Formato
Adobe PDF
|
2.69 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/10589/178852