No abstract

No abstract

This paper presents a new method for the identification of geometric design parameters of ISO 484 class propeller blades from scanned point cloud data. The method can be used for tolerance inspection and in-line measurement of manufactured blades, and for wear assessment and reverse engineering of propellers in service. The geometry of the propeller blades is specified by a set of blade sections stacked radially on concentric cylindrical surfaces. For each of these sections, the chord line and mean camber line is determined, and geometric design parameters such as the pitch, skew, chord length, camber, and thickness distributions are identified. The proposed method is a complete procedure for identifying the design parameters of marine propeller blades from point cloud data, and includes a novel method for the precise identification of the mean camber line based on Voronoi diagrams and Delaunay triangulation. The paper includes validation of the proposed method in experiments where reverse engineering is applied to a propeller blade of the KVLCC2 propeller, and in 3D scanning of a large, high-skew thruster blade where the results were compared to CMM measurements.

This paper presents a novel procedure for robotic scanning of weld grooves in large tubular T-joints. The procedure is designed to record the discrete weld groove scans using a commercially available line laser scanner which is attached to the robot end-effector. The advantage of the proposed algorithm is that it does not require any prior knowledge of the joint interface geometry, while only two initial scanning positions have to be specified. The position and orientation of the following scan are calculated using the data from two previous weld groove scans, so once initiated, the scanning process is fully autonomous. The procedure is a two-step algorithm consisting of the prediction and correction substeps, where the position and orientation of the sensor for the following scan are predicted and corrected. Such a procedure does not require frequent weld groove scanning for navigation along the groove. The performance of the proposed procedure is studied experimentally using an industrial-size T-joint specimen. Several cases of scanning motion parameters have been tested, and a discussion on the results is given.

No abstract