This paper describes the algorithms needed to allow automaticrecognition of components in three dimensions (3-D) using a novel activerange sensor based on liquid crystal technology, which is programmed toproject coded lighting sequences onto a pallet of workpieces within aflexible manufacturing cell. A range map with up to 1200×1200range points is constructed using a sequence of binary image capturesfrom a single robot head camera. Calibration of the vision sensor headis achieved using a grid of black squares on a white mat placed on thepellet prior to on-line operation and accuracy of 0.33% in heightmeasurement are achieved over a working area of 100×100 cm. Therange data is processed on a digital signal processing image analyserand construction solid geometry (CSG) surfaces are matched to the datasets to extract surface models of the components. The surface model isstored in an object database and used to recognise the position ofworkpieces in the robot coordinate frame
展开▼
