A positioning strategy for circular wireless charging coils based on low-dimensional feature fusion

For EV-oriented MCR-WPT systems, lateral displacement between the transmitting coil (TC) and the receiving coil (RC) alters the magnetic coupling condition and consequently weakens power transfer capability. Many reported localization schemes depend on extra sensing coils or dense sensor arrangements, which increase both hardware burden and signal-processing complexity. This paper proposes a two-dimensional positioning method for circular coil systems with a fixed vertical gap. A low-dimensional feature vector is constructed using the steady-state voltage at the secondary-side port and the outputs of two asymmetrically placed tunnel magnetoresistance sensors. Support vector regression is then employed to predict the radial offset distance and the sine and cosine of the offset angle, thereby enabling direct estimation of the relative position between the TC and RC. The proposed method is validated on 30 independent test points on a scaled-down experimental platform. The results show an average angle error of 6.4434° and an average Euclidean distance error of 0.1728 cm, demonstrating that the proposed method enables accurate two-dimensional positioning of circular wireless charging coils with low hardware overhead.