Experimental validation of ANFIS-based sensored field-oriented control for PMSM-driven AGVs

• It is commonly known that PID controllers are used with linear systems; however, because most controllers used to control PMSM drives operate within nonlinear systems, there are certain issues with control. • To address these issues, this paper presents an Adaptive Neural Fuzzy Inference System (ANFIS) for controlling this non-linear systems. • The algorithm will be implemented in a TMS320F28379D, and an inverter known as BOOSTXL-DRV8305EVM was used to regulate the PMSM drive and to acquire the current using a three shunt resistance sensor in an experiment to validate this controller. Permanent Magnet Synchronous Motors (PMSMs) are widely used in electric mobility and autonomous systems due to their high efficiency and dynamic performance. However, classical Field-Oriented Control (FOC) based on Proportional-Integral (PI) controllers often struggles with nonlinearities and parameter variations, leading to limited robustness and tracking accuracy. To address these issues, this paper proposes an intelligent control strategy based on an Adaptive Neuro-Fuzzy Inference System (ANFIS), replacing PI controllers in both the speed and current control loops. The PMSM is modeled in the d-q reference frame, and both control strategies PI-based FOC and ANFIS-based FOC are tested under identical simulation conditions. Results show that the ANFIS controller reduces the speed tracking error from 3–4% (PI) to less than 1%, shortens the settling time from 0.06 s to 0.04 s, and significantly minimizes torque ripple. These improvements demonstrate enhanced disturbance rejection and faster dynamic response. To validate the approach in real time, the proposed controller is implemented on a custom embedded board designed around a C2000 microcontroller, which integrates power electronics, current and voltage sensing, and CAN communication. The results confirm the effectiveness of the ANFIS strategy and the suitability of the developed hardware for high-performance applications such as Automated Guided Vehicles (AGVs).