[ad_1]
Loon’s AI can determine the optimal route for balloons much faster than the previous navigation system. It also does this more efficiently. Balloons can travel similar or greater distances than before with less power. Loon’s record-breaking flight time is 312 days – she set that benchmark earlier this year. Maybe the AI system can keep the balloons in the air even longer.
Dive
Loon and Google AI used simulations to train the RL model through trial and error before real-world testing began in Peru. The team then evaluated its capabilities directly against a human-designed system called StationSeeker with a 39-day test over the Pacific Ocean. The AI was able to keep balloons in target areas for longer periods of time while using less energy. This is important because it will help provide more consistent internet coverage to people in a given area.
StationSeeker tended to head straight for a specific target location, but it passed often and had to reverse its course. The AI system was more concerned with staying in the target area as passively as possible to save power when it needed it most. He also used complex maneuvers that Loon’s team had never seen before.
[ad_2]
Source link