The main challenge is that the parameters of the permanent magnet synchronous motor (PMSM) model are difficult to determine precisely and may vary during operation, which makes it difficult to ensure stable and high-performance control. Therefore, it is essential to employ algorithms such as observers and parameter adaptation schemes to compensate for these parameter variations. This paper aims to develop a speed controller for a permanent magnet synchronous motor that does not require prior motor model information, and to design a parameter optimization scheme for that controller. Specifically, the first contribution is the design of a model-free adaptive outer-loop speed controller that does not rely on prior motor parameter information, while the inner-loop current dynamics are regulated by an ABS-CC-based controller to ensure fast and stable current tracking under the Maximum Torque per Ampere (MTPA) constraint. The second contribution is an automatic tuning framework in which the parameters of both the outer speed controller and the inner ABS-CC current controller are optimally adjusted by an Improved Particle Swarm Optimization (IPSO) algorithm to enhance tracking accuracy and robustness against uncertainties. Finally, simulation studies are conducted to validate the theoretical developments, demonstrating high-performance speed and current control under the proposed approach.