Sensorless Sector Determination of Brushless DC Motors Using Maximum Likelihood Estimation

Brushless DC motors are widely used for their high power density and efficiency. However, sensorless control remains challenging due to the difficulty of accurate rotor position detection, especially at low speeds. This paper proposes a novel sensorless trapezoidal control method based on Maximum Likelihood Estimation (MLE) for rotor sector detection. Unlike conventional back-EMF zero-crossing techniques, the proposed method uses a statistical algorithm to generate a probability map from prior motor state data, enabling accurate rotor position estimation without sensors. The MLE method operates with a typical computation time of 50–100 (Formula presented.) s, offering a balanced tradeoff between speed and accuracy. It is significantly faster than Kalman filter-based approaches (200–1000 (Formula presented.) s) and comparable to observer-based methods (20–80 (Formula presented.) s), while being more robust than zero-crossing techniques (<5 (Formula presented.) s). This makes it a practical and cost-effective solution for applications demanding high efficiency and reliability, such as electric mobility systems.