The database can also be queried1 using a technique developed in Russia, known as the theory of inventive problem solving, or TRIZ. In essence, this is a set of rules that breaks down a problem into smaller parts, and those parts into particular functions that must be performed by components2 of the solution. Usually these functions are compared against a database of engineering patents, but Dr Vincent's team have substituted their database of biological patents instead. These are not patents in the conventional sense, of course, since the information will be available for use by anyone. By calling biomimetic tricks biological patents, the researchers are just emphasising that nature is, in effect, the patent holder3.
One way to use the system is to characterise an engineering problem in the form of a list of desirable features that the solution ought to have, and another list of undesirable4 features that it ought to avoid. The database is then searched for any biological patents that meet those criteria5. So, for example, searching for a means of defying gravity might produce a number of possible solutions taken from different flying creatures but described in engineering terms. If you want flight, you don't copy a bird, but you do copy the use of wings and aerofoils, says Dr Vincent.
He hopes that the database will store more than just blueprints6 for biological mechanisms7 that can be replicated8 using technology. Biomimetics can help with software, as well as hardware, as the robolobster built by Dr Ayers demonstrates. Its physical design and control systems are both biologically inspired. Most current robots, in contrast, are deterministically programmed. When building a robot, the designers must anticipate every contingency10 of the robot's environment and tell it how to respond in each case. Animal models, however, provide a plethora11 of proven solutions to real-world problems that could be useful in all sorts of applications. The set of behavioural acts that a lobster9 goes through when searching for food is exactly what one would want a robot to do to search for underwater mines, says Dr Ayers. It took nature millions of years of trial and error to evolve these behaviours, he says, so it would be silly not to take advantage of them.
Although Dr Vincent's database will not be capable of providing such specific results as control algorithms, it could help to identify natural systems and behaviours that might be useful to engineers. But it is still early days. So far the database contains only 2,500 patents. To make it really useful, Dr Vincent wants to collect ten times as many, a task for which he intends to ask the online community for help. Building a repository of nature's cleverest designs, he hopes, will eventually make it easier and quicker for engineers to steal and reuse them.
相关文章推荐
