As the core of electric propulsion in drones, the motor has higher requirements for its reliability and fault tolerance. Accurate acquisition of rotor position information is a prerequisite for a motor-driven drone’s system to operate stably. Traditional search coils can provide fault tolerance for position detection, but they cannot detect rotor position in the full speed range (stationary to rated speed). In order to make the search coils provide rotor position in the full speed range, this study proposes to inject an ultrahigh frequency (UHF) signal (50–100 kHz) into the search coils. By optimizing the self-inductance of the search coil, the mutual inductance between the search coil and the armature winding, the back electromotive force (BEMF) of the search coil, and the mutual inductance between the search coils, the structure of the UHF search coil designed in this paper is helpful to extract the UHF feedback signal. Finally, based on the mapping relationship between the self-inductance of the search coil and the rotor position, the rotor position of the motor can be detected in the full speed range. The novelty of the proposed work lies in the UHF search coil with zero mutual inductance coupling to the armature winding, small BEMF, and low interphase mutual inductance that can detect the rotor position in the full speed domain. Maxwell software is used to optimize the structure of the UHF search coil, and the feasibility of the design results is verified by co-simulation.