Brick-laying robots

Increasingly we are seeing repetitive tasks carried out by workers being replaced by robotics. This can even involve skilled workers, such as bricklayers. Mark Pavic of Western Australia, an aeronautic and mechanical engineer has with a colleague devised a brick-laying machine.

The story is told in a Gizmag story, Brick-laying robot can build a full-sized house in two days. The robot, Hadrian, can lay 1000 bricks an hour. Pavic's brother Mike is CEO of Fastbrick Robotics who intend to launch a commerical version in 2017.

It was as long ago as 2007 hat the world patent application for the concept was published, WO2007/076581. Below is the main drawing.

So, how could you have found such a patent document ? The classification is based on the idea of "manipulators" which could be run together with the word brick* to get a good list of relevant material. B25J9, programme-controlled manipulators, looks particularly attractive.

Once you know of a patent document, you can check to see what happened to it in specific jurisdictions. In this case US8166727 was granted protection in 2012. The pan-European granted document, EP1977058B, was published in 2014, and the documents to do with its allowance are given in the relevant European Register entry.

Anyone wondering at the time if the patents would be allowed could have looked at the prior art as listed by patent examiners -- this is given as "cited documents" on the left hand side of the bibliographic entry for the European A document.  22 are listed for that one. It might be thought logical that all prior art cited is given there, but no, each "also published as" document needs to be clicked on and a fresh window appears. Some countries, such as the USA, do not list cited patents at the initial A stage but only at the granted, B stage.

Just as you can find the prior art by clicking on "cited documents", you can find those that later referred back to the one you know of by clicking on "citing documents". This tells us that none cited the European; but for the US A and also the B document there are 9, and for the original WO document there are 6. You would have to compare the lists to figure out which were on both lists -- they are likely to be particularly significant.

Those with specialist knowledge of the area would then have to interpret the results, so long as those who understand the published patent documentation and its associated patent procedure can explain what to look out for and implications. Both are needed. When I worked in the area, it always surprised me how many inventors thought that they could do it all themselves, or just spend 20 minutes or so looking through the patents before committing to thousands of pounds of expenditure and huge amounts of time. Rather like carrying out brain surgery on yourself, perhaps ?