As Australia lurches from fiery catastrophe to fiery catastrophe, researchers have proposed the use of drones to provide better predictions of how bushfires may behave. Speaking to IT News, University of Melbourne researcher Dr Thomas Duff said fire prediction depends on data-gathering. As part of the Bushfire Cooperative …
I hope it can see through smoke
I can give you a clue on fire direction
- same direction as wind
- up hills
Can I have my $21 million now?
I can't wait for the first drone to crash into the bush and start a fire.
I also recall my probability lecturer talking about his work in Tasmania relating vegetation patterns to previous fires. This was circa 1980!
Right now I can hear an Elvis and 2 other water bombers dropping loads 10 KM away on a fire that kept a lot of us up last night. The wind blew the fire 40 kmh in 3 hours or so, but the fire did not always follow prevailing winds. It was heavily modified by hilly terrain and sometimes just does as it damn well pleases. My property had westerly winds while the flame track was clearly driven by SW winds as Bureau of Met forecast. Accurate ground truth of flame movement is essential for fire fighting as well as getting fire fighters out of the way. A local spot is known as Hells Pass for what happens there if a fire gets loose. Getting out of there requires time which is in short supply in an Oz summer.
Accurate immediate SitRep may also be useful for getting livestock out in time.
As a starter, perhaps the automated search and rescue drone that Canberra Linux Group members put together would be a great starting point. ITIRC it uses a Raspbery Pi and Open Source autopilot. However the Oz gov quangos are timid and prefer to buy some arctic rated equipment from a frozen hell hole than use locally designed and made stuff that works when the temperature goes over 30C. <sarcasm> I mean, it cant be good unless our good friends use it ? /<sarcasm> required for the intellectually impaired ;-)