Alphabet’s ambitious balloon project, called Loon, might soon make a speedy recovery from the problem that was hindering them previously, courtesy advancement in machine learning technology.
The project is a part the experimental division of the country called X and is now in its fourth year.
Earlier, the team of scientists at the helm of the project had proclaimed that the balloons would float around all over the world and were pondering over the technique to be used to keep them at a safe traveling distance from one another.
Now, the team says that it has found a way to make the balloons stay in a much more concentrated area and they will now circle around in a limited area instead of circumnavigating the globe.
The team credits their improved attitude control and navigation system for the same and stated that they can now toy with the idea of using of 20-30 balloons for an area instead of 200-300, as was the case before.
The balloons would now fly with the help of an artificial intelligence software instead of depending upon a set navigation plan that was used earlier.
Loon had earlier (September, last year) shared a video of its 98-day sojourn over Peru. After a few days, the balloons were found to be hovering the Yellowstone National Park in the US.
Even though the team has managed to get over this hurdle, there are several others still in their way. One amongst them is determining whether the cost of keeping the balloons afloat would be cheaper than installing infrastructure in the said area.