Towards high-performance force-controlled pneumatic actuator with long transmission lines: A novel nonlinear integral sliding surface and a new multimodal optimization algorithm
Loading...
Date
2018
Authors
Butt, Khurram Mahmood
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
A force-controlled pneumatic actuator with long connecting pneumatic tubes is a well-accepted solution to develop MRI-compatible force control applications. Such an actuator represents an uncertain, second-order, nonlinear system with input delay. The integral sliding mode control, because of guaranteed robustness against matched uncertainties throughout the system response, provides a favorable option to design a robust controller for the actuator. However, if the controller is based on a linear integral sliding surface, the response of the actuator overshoots, especially when there are large initial errors. Minimizing overshoot results in a smaller controller bandwidth and a slower system response. This thesis presents a novel nonlinear integral sliding surface to improve the transient response of the actuator. The proposed surface is a linear integral sliding surface augmented by a nonlinear function of tracking error and does not have a reaching phase when there are initial errors and even multiple steps in the desired trajectory. The surface enables the integral sliding mode controller to offer variable damping during the system response and minimizes the overshoot without compromising the controller bandwidth, rise, and settling times. The thesis also implements logically driven, time-efficient tuning of controller for desired transient response of the actuator. For this purpose, first a novel multimodal optimization algorithm is proposed which has better globalization and lower numerical cost as compared to the evolutionary, swarm and available globalized local search optimization methods. Next, the effectiveness of the proposed algorithm is examined for model-based and model-free tuning of a position controller for a servo pneumatic system whose performance has been reported in another doctoral thesis after rigorous trial-and-error tuning. On yielding better results, the same algorithm-based tuning is used to tune the force controllers having the linear and proposed nonlinear integral sliding surfaces for desired actuator response and performance comparison. Simulation studies and experimental tests conducted on a prototype MRI-compatible test rig show that the controller based on the proposed sliding surface successfully eliminates the overshoot without compromising the controller bandwidth, rise, and settling times. It also outperforms the controller having a linear integral sliding surface. The thesis also establishes the asymptotic stability of the proposed controller using Lyapunov’s stability criterion.
Description
Keywords
Force Control, Pneumatic actuator, Input delay, Integral sliding mode control, Nonlinear integral sliding surface, Controller tuning, Multimodal optimization
Citation
Butt, K., Rahman, R.A., Sepehri, N., & Filizadeh, S. (2017). Globalized and bounded Nelder-Mead algorithm with deterministic restarts for tuning controller parameters: Method and application. Optimal Control Applications and Methods, 38(6), 1042-1055.
Butt, K., & Sepehri, N. (2018). Model-free online tuning of controller parameters using a globalized local search algorithm. Optimal Control Applications and Methods, 39(5), 1750-1765.
Butt, K., & Sepehri, N. (2018). A nonlinear integral sliding surface to improve the transient response of a force-controlled pneumatic actuator with long transmission lines. Manuscript submitted for publication.
Butt, K., & Sepehri, N. (2018). Model-free online tuning of controller parameters using a globalized local search algorithm. Optimal Control Applications and Methods, 39(5), 1750-1765.
Butt, K., & Sepehri, N. (2018). A nonlinear integral sliding surface to improve the transient response of a force-controlled pneumatic actuator with long transmission lines. Manuscript submitted for publication.