from the now-try-performing-path-computation-in-real-time-with-on-board-cameras dept.
New Algorithm Flies Drones Faster than Human Racing Pilots:
The best human drone pilots are very good [...] and have so far always outperformed autonomous systems in drone racing. Now, a research group at the University of Zurich (UZH) has created an algorithm that can find the quickest trajectory to guide a quadrotor – a drone with four propellers – through a series of waypoints on a circuit. “Our drone beat the fastest lap of two world-class human pilots on an experimental race track”, says Davide Scaramuzza, who heads the Robotics and Perception Group at UZH and the Rescue Robotics Grand Challenge of the NCCR Robotics, which funded the research.
“The novelty of the algorithm is that it is the first to generate time-optimal trajectories that fully consider the drones’ limitations”, says Scaramuzza.
[...] The researchers had the algorithm and two human pilots fly the same quadrotor through a race circuit. They employed external cameras to precisely capture the motion of the drones and – in the case of the autonomous drone – to give real-time information to the algorithm on where the drone was at any moment. To ensure a fair comparison, the human pilots were given the opportunity to train on the circuit before the race. But the algorithm won: all its laps were faster than the human ones, and the performance was more consistent.
Journal Reference:
Philipp Foehn, Angel Romero, Davide Scaramuzza. Time-optimal planning for quadrotor waypoint flight [$], Science Robotics (DOI: 10.1126/scirobotics.abh1221)
(Score: 1, Interesting) by Anonymous Coward on Friday July 23 2021, @06:30AM (4 children)
When I was a kid I imagined a floating ball that hovered about 6 ft behind and above you that could warn you of things coming. Or anything really. Basically a genie that could solve problems, get things, do things, look out for you. Seems we are getting close to that point.
(Score: 1, Informative) by Anonymous Coward on Friday July 23 2021, @06:55AM
https://www.theverge.com/2017/7/17/15981250/japan-space-camera-drone-iss-int-ball [theverge.com]
(Score: 2) by fakefuck39 on Friday July 23 2021, @11:06AM
what we need is watch-sized drones with hours, not minutes of battery life, just auto-hovering above us, streaming video for our phone to record. I'm actually really surprised that's not a thing yet.
but we're close - as with many things, we need groundbreaking new battery tech.
https://www.dronezon.com/drone-reviews/best-follow-me-gps-mode-drone-technology-reviewed/ [dronezon.com]
(Score: 1, Funny) by Anonymous Coward on Friday July 23 2021, @12:02PM (1 child)
When I was a kid I used to have these nightmares of being stuck on an island somewhere and every time I tried to leave, a giant floating ball would pop out of nowhere and try to smother me.
Be seeing you.
(Score: 3, Insightful) by TrentDavey on Friday July 23 2021, @03:52PM
You are number 6.
(Score: 2) by MIRV888 on Friday July 23 2021, @03:13PM (7 children)
If an AI drone can outfly human pilots already, it's not going to be much of a sport.
Lot of that going around.
(Score: 2) by legont on Friday July 23 2021, @03:55PM (6 children)
I bet there were no hazards such as variable wind conditions. AIs can't predict local wind while human pilots routinely can.
"Wealth is the relentless enemy of understanding" - John Kenneth Galbraith.
(Score: 2) by MIRV888 on Friday July 23 2021, @05:50PM (5 children)
If the drone knows it's position in 3d dimensions, it can compensate in what would essentially be real time to us.
(Score: 2) by legont on Friday July 23 2021, @07:39PM (4 children)
This is now enough to win against a pilot who can "see" the wind before it comes using indications like ripples on water, leaves movement, and flying birds attitudes. Pilots can estimate future wind seconds or even minutes in advance.
"Wealth is the relentless enemy of understanding" - John Kenneth Galbraith.
(Score: 2) by FatPhil on Saturday July 24 2021, @03:39PM (3 children)
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 2) by legont on Tuesday July 27 2021, @03:23AM (2 children)
In aviation, reaction is nothing while prediction is everything. True for most other activities as well. When one reacts it's often already too late.
"Wealth is the relentless enemy of understanding" - John Kenneth Galbraith.
(Score: 2) by FatPhil on Sunday August 01 2021, @04:20PM (1 child)
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 2) by legont on Monday August 02 2021, @01:57PM
I don't think we lost this one. Let me repeat the original point. They probably run the test in ideal artificial conditions - not in a real wind, rain, snow and so on one. The article suspiciously does not mention it so I called it BS.
"Wealth is the relentless enemy of understanding" - John Kenneth Galbraith.
(Score: 2) by legont on Friday July 23 2021, @07:57PM (1 child)
Let's take another example. Say you are looking at a painting. Can you see wind on it? Where is it coming from? At what speed? Do it using different indications such as water, leaves and birds. Do you get the same result? If not, the painting is crap.
One can ask many more questions. What time is it? Season? Coordinates? Altitude? Is it consistent again? If a good artist painted it, it's all should be.
It'll be a long time before AIs can do this, while many trained professionals separate good painting from a bad one in a glance.
"Wealth is the relentless enemy of understanding" - John Kenneth Galbraith.
(Score: 2) by MIRV888 on Saturday July 24 2021, @06:52PM
I hear what you are saying. It's a perfectly valid argument. Knowledge of the updraft over ridge 'A' can allow the pilot to anticipate said updraft and compensate in advance. I just feel the power to weight ratio on drones in general would be able to compensate in most situations. A racing drone would have even better performance. I have had my Big6 (nickname she's 8.7 lbs) drone up in 20 mph gusting 40. She held her own quite well in gps lock mode. The nearest I came to getting thrown out of the air was a microburst. I see why pilots avoid them. Point being, drones perform well in variable wind conditions. That's why I believe the cpu will win.