This thesis investigates an artificial neural network (ANN)-based field-oriented control (FOC) for a surface-mounted and an interior-mounted permanent magnet synchronous machine (SPMSM and IPMSM). The ANN was trained by using Levenberg-Marquardt and forward accumulation through time algorithm. First, the thesis examines the fundamentals of motor parameters and two aforementioned vector controls, with training algorithms, in detail. Then, the background and various algorithms of Maximum Torque per Ampere (MTPA) and flux weakening (FW) control are undertaken while the following part epitomizes an off-the-shelf component-based electric vehicle (EV) model that is constructed using MATLAB SimPowerSystems and SimDriveline. The proposed control is validated in both simulation and hardware experiment and compared with a PI-based field-oriented control. First, for SPMSM, the results of simulation and hardware experiment show that the maximum operating speed of the proposed control is improved by 48% and 3.5% compared to the PI-based control. For IPMSM, the results show that the proposed control produces less d-axis current than the latter control. Moreover, the control is implemented and simulated in electric vehicle model, which is constructed using SimPowerSystems and SimDriveline library in Simulink by the author with off-the-shelf components. The results show that the proposed controller can be a potential replacement of the existing control schemes, such as PID, fuzzy logic, or others, and provides adequate traction control in EV application