This paper proposes a model-free predictive current control (MFPCC) method for permanent magnet synchronous motors (PMSMs) using a particle swarm optimisation-expanded Kalman filter (PSO-EKF) with an adaptive switching higher-order expanded state observer (ASHESO). Traditional MFPCC methods based on expanded state observers (ESO) rely on a hyperlocal model that eliminates the need for stator resistance and rotor magnetic chain parameters but still depends on stator inductance. Mismatched inductance parameters can hinder the ESO’s ability to accurately estimate time-varying perturbations, affecting performance and stability. The proposed method enhances the traditional second-order ESO by introducing ASHESO, which increases the number of dilated states to ensure asymptotic convergence of higher-order perturbations, improving estimation accuracy. Additionally, the expanded Kalman filter estimates inductance parameters, while particle swarm optimisation optimises noise covariance matrices (Q and R), enhancing parameter identification and model-free prediction. Compared to conventional ESO-based MFPCC, this method offers superior dynamic response, steady-state performance, and parameter robustness.