100 swarming drones from AsTec (now an Intel subsidiary) set a new world record. (The previous record was 30 [correction:the previous record was 50 (thanks Randy)] with a Pixhawk-powered swarm). We're just going to have to wait for the rains to clear and take that record back!
I would have a hard time saying that either one of these large swarm demos was significantly harder than the other (comparing the Naval Postgraduate School's swarm of 50 fixed-wing planes vs. the AscTec/Intel swarm of 100 multi-copters). The two projects had very different goals, but I'm sure they had to solve a lot of the same not-very-glamorous problems to be able to fly and control several dozen vehicles at the same time.
Because the fixed-wing swarm was launched one airplane at a time, they had to achieve and sustain a launch rate that let them get the last plane in the air while the first plane still had enough battery to complete their actual test. Their test was focused on an intentionally loose method of coordination between members of the swarm to study emergent behaviors and validate their simulations.
The multi-copter vehicles could all be launched at the same time, but they needed to execute several minutes of precision 3D flight formations which were synchronized with the performance of a live orchestra. In the dark.
Both are big achievements and both groups should be very proud. I don't think either will be duplicated soon.
I think it is interesting that each of these very different projects ended up choosing to manage multiple "sub-swarms' of 25 vehicles. I wonder if both teams hit the same limits on GCS radio connections or wifi bandwidth or something?
This is an impressive display, truly incredible. The need for sub-swarms is to enable the communication links to remain in tact with each drone. When operating 25+ drones on a defined band, it saturates quickly and makes it difficult to maintain a continuous link, by dividing the 900MHz band into smaller bands, x number of drones can be applied to each band thereby enabling the connection links to be maintained.
Also, I work for Queen B Robotics based in Berkeley, we build swarm systems for drones. If we could get our hands on a big batch of Iris+ drones, or PIxhawks, we would be willing to put the record up to 150...
@Chris this is DIY Drones, build your 150 aircraft. Probably cost you between $300-$400 for an autonomous aircraft. So for about $60K, you would have the hardware you need for 150 dots in the sky.
I think distributing this idea and flying 1000's of autonomous aircraft/robots/whatever at the same time/synchronized globally allows everyone to participate in an event driven by community and not corporate acquisition. Folks in Spain, Netherlands, Germany, France, US are already on board. This could be a lot of fun and bring together all the projects related to autonomous machines globally. http://loverobot.org is the place holder for the idea as we start to flush out the details. Maybe Queen B Robotics with your existing fleets want to coordinate something in the Berkley area? T-minus 167 days and counting. :)
@Tom, I am sure they would be happy to take our money.
@earthpatrol, we would be happy to help coordinate this event and make something like this happen. Let's make this happen!
Join DIY Drones
Welcome toDIY Drones
Sign Upor Sign In
Or sign in with:
© 2019 Created by Chris Anderson.
Report an Issue |
Terms of Service
Please check your browser settings or contact your system administrator.