CSIRO take up the lead....CSIRO take up the lead....
From Fantasy to Possibility
The CSIRO Division of Textile Industry in Geelong occupies a typical laboratory building in salmon brick and aluminium, a true product of the sixties. My fellow reviewers and I were met by David Henshaw with his neatly trimmed moustache, silvery hair and white coat, looking the quintessential CSIRO scientist. He took us downstairs to a basement room neatly panelled in varnished wood. In the centre of the room was a device that could only be described as a contraption. A shearing handpiece hung from the end of it, supported by a series of sloping metal tubes pivoted on a large frame about the height of a man. Wires of all colours draped across it, secured with sticky tape and leading to large rather old-fashioned looking electrical cabinets behind the machine. In amongst this was a coiled orange air hose, supplying compressed air to run the handpiece motor.
Laid out around the room were memorabilia of David's
early experiments; photographs, charts, and dismembered pieces of equipment.
There was a temporary hush as David turned his machine on. As soon as he touched the comb, it suddenly came to life with a jittering and rattling noise, the rods whirling this way and that, the handpiece seeming to dance on his hand. Two sensing pads just behind and underneath the comb were reacting to David's fingers; gentle pressure was sufficient to move the comb away. David moistened his fingers and demonstrated the same effect when he touched the points of the comb.
A sheep made a quiet entry to the room, legs splayed apart, its hooves slithering on the polished vinyl floor, its body tightly secured between the knees of a white-coated lab technician. Two or three people helped him heave it up so that it was sitting astride a beam carried by a small trolley on castors. The legs were secured into four clips and the head was strapped down over the front end of the beam.
The chatting groups of people around the room quietened expectantly as the trolley was pushed into position in front of the machine. An electric handpiece was used to shear a wide swathe by hand down the backbone from the head to the tail. I was struck by the contrast between the white fluffy fleece on this animal, and the hot greasy grey wool of the sheep at Muresk full of dust, grit and grass seeds.
The trolley was secured to the front of the machine with the shears pointing down just above the backbone. With a piercing growl the air motor came to life and the points of the comb descended into the shorn area along the backbone of the sheep. They landed gently on the skin and danced down the side of the sheep, lifting out a neat lump of shorn wool.
As at Muresk, anxious arms pulled back the wool to reveal
the neatly shorn and intact skin beneath. The trolley was moved one space
to the side and registered once again, against a notched piece of wood so
that the next shearing blow would be just alongside the first. Soon a dozen
blows had been shorn and the upper half of the side of the sheep was cleanly
shorn from one end to the other. The sheep was tilted on its beam and another
row of blows shorn to complete the shearing test. The entire side had been
slowly but cleanly shorn with one noticeable ridge where the cutter had
neatly climbed over a wart leaving deeper stubble on either side. An amazing
demonstration!
In contrast to the quiet murmuring around the difficulties of the PATS machine at Muresk the chatter was jovial and optimistic but still restrained as another sheep was mounted on the trolley.
This was far more experimental than the PATS machine. Almost 'Heath Robinson'with pieces of string tying it together, but it worked. (Heath Robinson was a whimsical cartoonist who drew elaborate, but makeshift machines for achieving unlikely tasks such as picking up safety razor blades. His machines were often driven by long strings running over pulleys, turning levers or tripping over buckets.)
Every now and again the cutter bucked and rocked forwards and backwards but somehow there was no more serious effect on shearing than slight ripples in the stubble. Every now and again the cutter refused to start. No doubt a weakness of the makeshift pneumatic drive.
The second sheep had a noticeable hollow in its side and the performance of the machine in climbing out on the ridge of the rib cage and then diving into the hollow of the stomach was quite astounding. It was a convincing demonstration of the effectiveness of feedback control. Here was demonstrable evidence of the possibility of sheep shearing robots.
Once again, as in the case of the PATS machine,
David had chosen a hard wired controller. As the external appearance of
the machine suggested, the controller was even more primitive, consisting
of half a dozen or so electronic amplifier circuits. Yet in spite of its
crude construction and simple design it came out a winner on performance.
In reality, both the CSIRO and PATS machines had clear strengths and weaknesses and the talk at the moment was on how the expertise which had designed and constructed them could be brought together. David had avoided the problem of supporting the sheep for shearing any area apart from the side. Even though the PATS machine had its deficiencies at least it was possible to access most parts of the sheep.
David's control system could clearly do with a great deal of improvement to reduce the bucking and rocking motion of the cutter. There were unanswered questions about the effect on the resistance sensing of wool contamination and moisture - effects which would not worry the PATS machine.
I had been impressed with the performance of the PATS machine in Muresk but now I was excited. Of the two, David's was the one which had crossed my intuitive boundary between fantasy and possibility. There was a real shearing robot in my imagination; free to move in harmony with the sheep, able to follow complex curves just as the arm of a shearer does. David's machine, with its short, cramped movements, constrained to the straight line of its supporting frame, was just a starting point. It seemed natural to me to extend the idea into the six dimensions of free space.
I had turned the corner - first a sceptic, now excited and convinced that it could be done.
In retrospect, the report which I wrote in the following days was both understated and over-optimistic. I concluded that both the CSIRO and PATS machines were too primitive to develop further, but a composite design had a good chance of success. I suggested that a specification for an automatic shearing system should be drawn up so that intensive development could proceed with a realistic budget and time scale. Collaborative research was needed; it wouldn't be practical to bring all the expertise and supporting facilities together. I advocated coordination between the different organizations involved. I warned that it might be hard to establish a satisfactory working relationship between the participants. At least I was right in that respect!
|Tour Page|James Trevelyan's
Page| UWA Home page|
Web site design: James Trevelyan jamest@mech.uwa.edu.au
September 1997