Slashdot Mirror
NASA Boosts AI For Planetary Rovers
transcendent writes "According to Space Daily, NASA is working on increasing the ability of future rover's AI. From the article: 'It now takes the human-robot teams on two worlds several days to achieve each of many individual objectives... A robot equipped with AI, on the other hand, could make an evaluation on the spot, achieve its mission faster and explore more'. Sounds like a good idea, but the article continues, 'Today's technology can make a rover as smart as a cockroach, but the problem is it's an unproven technology'. Another article about autonomous rovers being developed by Carnegie Mellon University is here."
The more famous quotation (which I suspect is the root of the 'cockroack' descriptor) is: "Robots today have the collective knowledge and wisdom of a cockroach... a retarded cockroach... a lobotomized, retarded cockroach." -Dr. Michio Kaku
People wanting to get some feel of AI, take a look atl ;)
http://www.ai-junkie.com/ann/evolved/nnt1.htm
It's a small app that automagically learns minesweepers to
pick up mines
We're doing similar work at the University of Sunderland. See http://www.his.sunderland.ac.uk/. My specialty is 'batbots' - sonar-controlled robots that exhibit sensorimotor integration.
Guess what? A neural network is a simple nonlinear function. Period. Training such a thing is nothing more than estimating its parameters by minimizing some (usually quadratic) cost criterion. When you put something in, you merely evaluate a rather simple nonlinear function. There is no intelligence involved!
It's a while since I did neural nets and university, but IIRC a 'nueron' in a nueral net lacks nothing from its biolgical equivlent. Neurons in living creatures either fire or not based on the their imputs, those neurons connecting to them that are firing, and the weight they have.
The problem is in building the networks, biological networks are complex, with mutliple sub-networks parrallel processing, with those sub-networks interacting. Humans can't yet build networks that can do anything like this.
Still, if an artificial neural net is a simple, non-linear function, then all the more complex networks in biology are 'just' collections of such functions. Complex behaviour arriving from lots of simple, but interacting, components.