Akademik Veri Yönetim Sistemi
Avrupa Bilim ve Teknoloji Dergisi, cilt.23, ss.55-62, 2021 (Hakemli Dergi)
Pneumatic Artificial Muscles (PAM) are soft actuators with advantages such as high force to weight ratio, flexible structure and low cost. Pneumatic Artificial Muscles have inherent compliance that makes them feasible for exoskeletons and rehabilitation robots. However, their inherent nonlinear characteristics yield difficulties in modelling and control actions, which is an important factor restricting use of PAM. The compliance of PAM is associated with nonlinearity, hysteresis, and time varying characteristics, which makes it more difficult to model the dynamics and operation with model based high-performance controllers. Although there are many studies to overcome the modelling issue such as virtual work , empirical and phenomenological models, they are either much complicated or very approximate ones as a variable stiffness spring for model with nonlinear input-output relationship. Based on the analysis of well known previous modeling works in our PAM test bed, it has been observed that efficacy of the those methods are limited for representing the physical behaviour of PAM and thus there is still requirement for simple and effective models . In this work, apart from previous modeling approaches, the behaviour of PAM is foreseen as an integrated response to pressure input, which results in simultaneous force and muscle length change. Therefore, standard direct input-output identification methods are not suitable for modelling that behaviour. An inverse modeling approach is proposed in order to utilize it in control applications. The black box model is implemented by an Artificial Neural Network (ANN) structure using the experimental data collected from the PAM test bed. According to implementation results, an ANN based inverse model has yielded satisfactory performance deducing that it could be a simple and effective solution for PAM modelling and control .
Keywords: Soft Actuators, Pneumatic Artificial Muscles, Inverse Modeling, Artificial Neural Network Based Modelling.