Since the manufacturing process of aircraft panel parts resquires accurately extracting the machining allowance， this paper proposes a repair method of panel parts based on the measured point cloud data. First， a feature point detection algorithm based on normal difference is designed to extract the initial feature points of the reference panels. Second， the iterative shrinkage optimization model is established to perform shrinkage optimization on the initial feature points to obtain the final feature points of the reference panels. Finally， registration is performed according to the tooling positioning holes， and after the feature points of the reference panel are mapped to the panel to be repaired， feature points are connected in a certain order to obtain the processing path of the panel to be repaired. Processing is carried out according to the repairing path and the gap of the panels is detected after the assemble. Experimental result show that this method can accurately extract the processing path of the panel to be repaired and after processing verification， and the gap between the wall panels is less than 1.2 mm.