مدل اولیه برای اسکلت خارجی اندام تحتانی
Abstract
I. Introduction
II. Dynamic Model
III. Motor Model
IV. Design of IMRAC
V. Lyapunov Stability
Authors
Figures
References
Abstract
In this paper, a Proportional-Integral-Derivative (PID) controller tuning scheme by Initialized Model Reference Adaptive Control (IMRAC) for a Lower Limb Exoskeleton (LLE) is presented. Mathematical expression of the LLE structure is determined using Lagrangian and Kirchoff’s equations. The transfer function of the structure based on the physical features of the links, and actuators is established under Range of Motion (RoM) condition. The PID controller of the LLE is tuned in a closed-loop control system using Ziegler-Nichols (Z-N) for initializing parameters of IMRAC. Adjustment mechanism is a gradient based method for real-time adaptation of tuned PID controller. A Lyapunov function has been applied to confirm the stability of IMRAC. The proposed IMRAC shows faster convergence in comparison with conventional non-initialized model reference adaptive control. It can be ascertained the proposed tuning scheme is applicable for real-time tuning of PID controller of LLE.
Introduction
In recent years, the demands for exoskeleton as a rehabilitation device have been increased, because of the growing population of elderly people and brain injuries such as stroke and Spinal Cord Injury (SCI) [1]–[3]. Therefore, development of the exoskeleton has been focused in assisting patients who lost their muscle stamina to recover their mobility function [4]–[6]. Lower Limb Exoskeleton (LLE) is a type of rehabilitation wearable robots, which is worn in parallel with patients’ lower limb. The main application of LLE is to help physiotherapist for gait training [7]–[9]. Thus, designing a robust controller that can work under different conditions is essential for optimizing the LLE performance. In this paper, Proportional-Integral-Derivative (PID) is developed as a controller in an adaptive control system. Combination of adaptive control and PID is used in several works [10], [11]. Aboud et al. [12] proposed multiple model adaptive control schemes, in which PID controller is selected as the controller for a mechatronic suspension system. Han et al. [13] developed a model-free based adaptive consists of intelligent Proportional-Integral (PI) controller, time-delay estimation, and adaptive sliding mode compensator for a 12 DoF LLE. Their control strategy in a simulated model was verified, and the stability was validated via Lyapunov theory. Wang et al. [14] combined fuzzy self-adaptive and variable-speed integral PID and presented fuzzy adaptive PID hybrid control strategy for a Multi-rotor Unmanned Aerial Vehicle (MUAV).