A Versatile Computer-Controlled Assembly System

A versatile assembly system, using TV cameras and oomputer-controlled arm and moving table, is described. It makes almple assemblies such aa a peg and rings and a toy car. It separates parts from a heap, recognising them with an overhead camera, then assembles them by feel. It can be instructed to perform a new task with different parte by spending an hour showing it the parts and a day or two programming the assembly manipulations. A hierarchical description of parts, views, outlines etc. is used to construct models, and a structure matching algorithm is used in recognition. 1, Introduction A computer-controlled versatile assembly system has been programmed during the past 12 months using, the Edinburgh hand-eye hardware (Barrow aid Crawford) The equipment (Fig. 1) consists of a moveable table, a mechanical hand with sensors and rotating palms, and two TV cameras, all conneoted via an 8X Honeywell 316 to a 128K time-shared ICL 4130 running POP-2 programs. Several other programs are running on thia equipment, including a program for recognising irregular objects and one which packs arbitrarily shaped objects into a box (Mlchie et al). The program described here is our most ambitious effort. It is capable of assembling a variety of structures, and much of our effort has been spent in enabling the machine to acquire descriptions of the parts for itself using an overhead TV camera. Related work has been carried out at Hitachi (EJiri et al), at MIT (Winston 45) and at Stanford University (Feldman). The Hitachi program could build a variety of simple structures of blocks from line drawings of the structure, the HIT programs can learn concepts about structures and copy an arbitrary structure of simple blocks given spare parts, and a recent Stanford program can assemble a simple auto­ mobile water pump using preprogrammed hand manipulations. 2. The TAsk A number of parta are placed by the operator in a heap on the table (Fig. 2, peg and rings). The machine's task is to separate the parts and recognise them, then to assemble them into some predetermined configuration (Fig. 3). Figs. 4 and 5 show another example, a toy oar. We are currently thinking in terms of up to a dozen parts with outlines described by up to twenty or so straight or ourved segments from any one view, possibly with some holes of similar complexity. In order to explore the capabilities of a …