Angle and Position Control of a High-Friction Pendulum on Cart – Comparison of SMC and AFSMC Approaches

The study focuses on controlling the angle and position of a high-friction Inverted Pendulum on ‎a moving Cart (IPC) system. An experimental setup with a low-quality, high-friction gearbox is ‎built to make the problem more challenging. The friction force is measured and found to be ‎dependent on the Cart position. A simple position-dependent friction curve is fitted to the ‎experimental measurement and added to the dynamic model of the plant for simulation ‎purposes. Since the IPC dynamic equation is not input-output linearizable, an approximate ‎feedback linearization method is employed, followed by a Sliding Mode Control (SMC) ‎approach. A Direct-Adaptive Fuzzy Sliding Mode Control (AFSMC) approach is then tailored ‎to mimic the feedback linearization part using an adaptive fuzzy engine, reducing the model-‎based part of the control. The uncertainty bound is estimated online and used in the switching ‎part of the controller to reduce control input chatter. Both SMC and the less model-dependent ‎AFSMC are implemented in simulations and practical implementations. While both methods ‎perform well in the nominal case, the superior performance of AFSMC is revealed when ‎intentionally induced uncertainty and noise are applied to the model and to the Cart position ‎sensor, respectively.‎