ABSTRACT The detection and parameter estimation of radar manoeuvring targets has attracted more and more attention in recent years. Due to the high manoeuvrability of the target during radar observation, range migration and Doppler migration will occur and severely deteriorate the coherent integration performance of the target echo. In this letter, an efficient manoueuvering target detection method is proposed, and the target's multi‐dimensional parameters are estimated with a coarse‐to‐fine strategy, which combines the Newton–Raphson‐based optimiser and sparse Bayesian learning. Compared with the existing methods, the proposed method achieves a better balance between computational efficiency and estimation accuracy. The simulation results are provided to validate the effectiveness of the proposed method.