Enhancing Underwater Robot Manipulators with a Hybrid Sliding Mode Controller and Neural-Fuzzy Algorithm

Underwater robot arms work in fast‑changing, uncertain, and noisy conditions, so following a precise path is hard. Sliding‑mode control (SMC) can keep the system stable even when parameters change, but it normally needs an exact mathematical model and it creates high‑frequency “chattering” when the state is close to the sliding surface. To overcome these problems, this work proposes a hybrid controller that mixes three techniques: – **Radial‑Basis‑Function Neural Network (RBFNN)** – learns and approximates the complex, unmodelled dynamics of a 3‑DOF manipulator. – **Fuzzy‑logic scaling** – adjusts the amplitude of the SMC law on the fly, which reduces the chattering effect. – **Linear Parameter‑Varying (LPV) model with a Kalman filter**

– filters out environmental noise and further improves overall stability. The combined system’s stability is proved with Lyapunov theory. Simulations show that the hybrid controller produces smooth, well‑controlled motions, with noticeably less chattering and lower sensitivity to noise. Putting this controller on a real underwater robot will require careful tuning of the fuzzy membership functions and the neural‑network weights. Doing so can be computationally demanding for the limited hardware typically available on underwater platforms.

Message: Hi FE, I need help with “Enhancing Underwater Robot Manipulators with a Hybrid Sliding Mode Controller and Neural-Fuzzy Algorithm” in “Electrical Engineering”