Saildrone makes history

Just because it functions doesn't make it desirable. The concept sounds goofy to me.

 

Oh sure, I get what you're saying. I wouldn't be keen on hordes of these things buzzing around all over the place either. It is fairly small and slow though, so any collision isn't likely to be horrific for any manned craft, and should be avoidable anyway.

 

It also is dangerous to surface navigation. Who cares that it has Ais...95 percent of the world fleet of surface vessels is not equipped with ais.

 

TBH I think too many of these drones would primarily be dangerous to each other, not to craft of any significant size. I can also see the arguments in favour of them for research purposes.

 

Ocean.

Um, I don't want to burst your bubble but the Pacific ocean ends at Australia, not Hawaii !

 

haha, true but its still a massive voyage in ocean condition. I think its impressive.

 

True

True, it's on its way! No bubbles burst for them. It has broken ALL autonomous sailing records. Not bad.

 

?

Have you ever sailed? It seams that you have very little idea about anything.

 

Ive spent my whole life sailing. two dozen trips across the atlantic in small craft. Typically 150 days on watch every year for the past 40 years


How about you wise *** ?

 

Well done.

 

Any vessel navigating at sea has a responsibility to obey the COLREGS. If I'm out there with my cruising trimaran, I'm not going to show up on AIS. It's my responsibility to avoid obstacles like floating arrays. But if I'm on starboard tack and another sailcraft is approaching on port tack, it's their responsibility to avoid me, because I'm on the right and am the stand-on boat. They are the give-way boat.

Saildrone is 19 ft long and has an average speed of 3 - 5 kt with a top speed of 14 kt. If Saildrone and my boat are on a collision course and each doing 10 kt, it could really do some damage. It would not be acceptable for, say, a 19 ft racing trimaran to proceed at 10 kt without any regard for other traffic. We would expect the racing trimaran on port tack to duck a cruising boat on starboard tack. The same thing goes for Saildrone.

Collision avoidance is THE critical technology that needs to be developed for autonomous craft of all types, whether sailing or powered. Without being able to maintain an effective lookout, it is a hazard to navigation. If it is necessary to develop the aerodynamic and hydrodynamic configuration, onboard systems, etc., without having an autonomous collision avoidance capability, then the craft should be escorted by a tender that is capable of maintaining a lookout for it.

Simply loosing it alone on the high seas is an irresponsible act that should not be encouraged. And I say that despite being a supporter of autonomous craft, a controls engineer, and a personal friend of the developer of Saildrone. The technology for autonomous collision avoidance of non-cooperating vessels is not that far away. It's as much a matter of will as it is of advancing the state of the art.

 

Um, no, it doesn't. First of all, it has not crossed an ocean, only made a long passage.

Second, it's not the first autonomous craft to cross an ocean or even the Pacific. Wave Glider crossed the Pacific on Dec 5, 2012.

 

We'll Tom. Quote was first autonomous sail boat. Not craft. Everyone knows that Wave Glider floated to Hawaii in over 3 months. Not wind powered.
You do have good points about safety and collision avoidance.
It's my understanding that,at the moment, drones or unmanned vessels are catorgorized as buoys by the coastguard so they do not need to obey COLREGS right now. This is all very new!
The only way to test new technology like this is to test on the open seas, in the real world.
You are an innovative guy, I'm sure you would understand testing a platform in real conditions.
These are a smart bunch of guys and I'm sure the issues you rightly bought up are being worked on and developed.
Looks like Richard Jenkins is answering questions on Sailing Anarchy tomorrow so maybe we can get some answers!

 

I don't see why detection, avoidance and response can't be automated. Mars landers detect the surface, modify their tracks and do much more. If something goes wrong with redundant systems big brother plucks it out of the sea before warning signals are depleted, or the down range officer scuttles it.

A similar problem is being addressed by the FAA. Currently thousands of drones are operating most without detection or avoidance algorithms. The recreational drone market (UAV) market has exploded and new rules are being developed to provide for their coexistence with manned vehicles http://www.faa.gov/about/initiatives/uas/

Dino

 

Sorry the air craft anology wont work. Every aircraft or. Drone is on a known route, these aircraft on known routes are tracked and regulated by a gigantic FAA infrastructure with thousands on employees.