Sliding Mode Control Made Smarter: A Computational Intelligence Perspective

Sliding mode control (SMC) is a well-known control method that has been widely studied and applied for more than 50 years since its inception in the late 1950s [1], [2]. While its simplicity in methodology and robustness against certain uncertainties and disturbances are celebrated, its shortcomings, such as chattering and brutalness of control forces, are also well documented. Computational intelligence (CI) techniques, such as neural networks (NNs), fuzzy systems (FSs), and evolutionary computation (EC), can provide means to help overcome the shortcomings. This article introduces the basic concepts and principles of SMC, shows how CI techniques can be tailored to make SMC smarter (in the sense of reducing chattering and control brutalness), and speculates on the future of SMC.