This paper outlines linear and nonlinear systems, presenting some key features of each while pointing out the major problems brought by the nonlinearities into control applications. Linear systems exhibit proportional and additive relationships and are comparatively easier to model and control. On the contrary, the nonlinear systems possess complex and unpredictable patterns, thus often forcing their simplification. Linearization, performed via Jacobian-based approximations and feedback linearization, is very important in allowing the application of linear control strategies to be employed for nonlinear systems. The paper presents proportional-integral-derivative (PID) control and Computed torque control (CTC) as the major control methods for manipulation, showing the corresponding design principles and also discusses other control techniques in brief.
Garg et al. (Wed,) studied this question.