Journal of Manufacturing Systems Volume 13/No. 3 Sensor-Based Solution to Contiguous and Overlapping Parts in Vibratory Bowl Feeders Gary P. Maul and Nebojsa I. Jaksic, Ohio State University, Columbus Abstract This paper discusses a vibratory bowl feeder with a computer-based sensor system that can recognize correctly oriented parts even when some are contigu- ous or overlapping. Hardware consists of a vibratory bowl feeder, sensor system, control computer, and software development computer. The sensor system is a linear row of film optics in the side wall and track of the bowl. This configuration creates a 3-D silhouette of the part. An algorithm is developed to recognize parts in the orientation specified by the user. Test results show that the control program functions very well. Keywords: Vibratory Bowls, Low-Cost Vision Sys- tems, Sensors, Part Feeders Introduction A part feeder can be defined as a device that receives a number of randomly oriented parts at its input and delivers parts in a unique orientation at its output. Part feeders can be divided into vibratory and nonvibratory types. ~ The most versatile and widely used type of part feeder is the vibratory bowl feeder) In a vibratory bowl feeder, parts move along the track of the bowl. Orientation is performed by mechanical devices mounted along the track. By adjusting these devices, the vibratory bowl feeder can be configured to feed a part in a specific orientation. The feeder must be taken off-line and retooled whenever the part or its desired orientation is changed. Flexibility of vibratory bowl feeders can be increased by replacing the mechanical devices with a sensor system. Parts first settle into one of a number of stable orientations that make it easier for the part to be recognized than in a random orientation. Traditional camera vision systems are expensive, which limits their use in feeding applications in industry. 3 Image processing algorithms for low- resolution systems are usually simple thresholding or edge-detection algorithms. Pattern recognition algorithms deal with tasks such as recognition of single or multiple objects, nontouching or touching parts, and nonoverlapping or occluded parts. Cronshaw et al. 4 developed a flexible assembly module using a vibratory bowl feeder. The part is fed by the bowl in any one of its stable orientations onto a transferring belt. When the part reaches the end of the belt, a pusher moves it past an inspection station. The station consists of a lamp, a prism to reflect light sideways onto the side wall of the track, two thin strips of optical fibers embedded in the track--one across each wall--and a semiconductor line-scan camera. The image of the part is formed by rapid scans of the camera. The pusher is used to ensure that the part travels at a constant speed while passing the inspection station. A 2-D binary image is built by the camera. Next a microcomputer analyzes the image, decides whether the part is scrap or not, and whether it is in the correct orientation. In other work, a programmable silhouette recog- nizer (PSR) was developed, s A simple sensor sys- tem is mounted in the track of the bowl at the feeder outlet. A grid formed by light sensors functions as a camera. A light source is placed above the sensor grid. The system uses a small microcomputer to compare and recognize part silhouettes. An air jet mounted in the bowl wall at the outlet returns parts with the wrong orientation back into the bowl and allows correctly oriented parts to pass. Light and dark information is converted into digital signals by 16 phototransistors. Each workpiece is represented by a unique 16-bit digital pattern. An enhancement 190