Skip to content

Sliding Mode Control Theory

Sliding Mode Control (SMC) is a robust control strategy that belongs to the class of Variable Structure Control (VSC). It is particularly effective for systems with high-frequency dynamics, non-linearities, and significant model uncertainties or external disturbances. Unlike traditional PID controllers that adjust parameters to achieve performance, SMC forces the system's state onto a predefined surface in the state space and keeps it there by switching the control signal at high frequency.

The core idea of SMC is to define a "sliding surface" where the system's error dynamics are stable and have desirable properties. Once the system's state is on this surface, it "slides" toward the origin (equilibrium), regardless of any external disturbances, provided they are bounded. This property, known as invariance, is what makes SMC exceptionally robust compared to linear control techniques like LQR or PID.

In practice, SMC is implemented in two main phases: the reaching phase and the sliding phase. During the reaching phase, the control input is designed to drive the system state from an arbitrary initial position toward the sliding surface in finite time. Once the state hits the surface, the sliding phase begins, and the control law ensures that the state remains on the surface for all future time.

Key Concepts

1. Sliding Surface (\(s\))

The sliding surface is typically defined as a function of the tracking error \(e\) and its derivatives. For a linear surface, we might have: $\(s = \dot{e} + ce = 0\)$ When \(s=0\), the error dynamics \(\dot{e} = -ce\) ensure that the error \(e\) decays exponentially to zero.

2. The Reaching Phase

To ensure the system reaches the surface, we design a control law such that the Lyapunov-like condition \(s\dot{s} < 0\) is satisfied. A common reaching law is the constant rate reaching law: $\(\dot{s} = -k \cdot \text{sgn}(s)\)$ where \(\text{sgn}(s)\) is the signum function.

3. The Chattering Problem

A major challenge in SMC is chattering—high-frequency oscillations of the control signal and the system state around the sliding surface. These oscillations are caused by the ideal switching nature of the signum function and can lead to mechanical wear or instability in real-world systems.

OpenSMC provides several solutions to mitigate chattering: - Boundary Layer Approach: Replacing the signum function with a continuous approximation like tanh or sat. - Higher-Order SMC (HOSMC): Moving the switching action to higher-order derivatives of the control signal (e.g., Super-Twisting algorithm). - Adaptive SMC: Dynamically adjusting controller gains to the minimum required level.

Further Reading

  • Utkin, V. I. (1992). Sliding Modes in Control and Optimization. Springer-Verlag.
  • Edwards, C., & Spurgeon, S. (1998). Sliding Mode Control: Theory and Applications. Taylor & Francis.
  • Shtessel, Y., Edwards, C., Fridman, L., & Levant, A. (2014). Sliding Mode Control and Observation. Birkhauser.