FAA regs preventing Amazon's Move into Drone Delivery Systems".

[Don Testme]

From dashcam footage on YouTube that I've seen of the Prime trucks, the cost for letterboxes demolished by those low skilled drivers must cost Amazon quite a bit though

A Lot of people use their own delivery trucks or vans to deliver Amazon products. I often have plane white vans pull up to deliver amazon products to my house. I asked them,"Do you own that van?" they say yes.

 

[Don Testme]

HAS ANYONE TRIED TO RUN ONE FULLY 360 OBSTACLE COLLISION DRONE INTENTIONALLY INTO ANOTHER FULLY 360 OBSTACLE AVOIDANCE DRONE?

 

[old man mavic]

@Don Testme ,well i am sure we will all await the results of your test at some point in the future

 

[Rchawks]

 

[Don Testme]

@Don Testme ,well i am sure we will all await the results of your test at some point in the future

Yes, I was going to do it today. I was going to pit an EVO Lit+ against a Mavic 3. Because the Evo Lite+ only has forward and back sensors, I was going to see if a head on collision might occur. But it'll have to wait. I haven't used tge Evo Lite+ in so long the R/C and Display batteries were dead. Maybe tomorrow.

No worries, if they get too close I'll stop the approach.

 

[old man mavic]

@Don Testme Hi fellow flyer thanks for the follow ,maybe fly the drones towards the rear of each other ,could help with gimbal damage ,in the event they do meet up its the most delicate part of any drone

 

[Don Testme]

@Don Testme Hi fellow flyer thanks for the follow ,maybe fly the drones towards the rear of each other ,could help with gimbal damage ,in the event they do meet up its the most delicate part of any drone

The above link is a video showing me trying to get my mavic 3 and evo lite+ to hit eachother, at least where they have obstacle sensors. I used the front ones for this experiment. This experiment wasn't a test of high speed approach, it was a low speed test. I wanted to see if the drones would see each other and stop. THEY STOPPED EACH TIME. Or, they tried going around each other. I could not get them to collide no matter how hard I tried.

I might make a higher speed collision risk video, but I don't know if im brave enough. But I expect they will avoid eachother or break. Breaking would be bad though. I would have to set them both to AVOID obstacles.

Why am I posting this video? It shows that even if the skies were cluttered with drones, especially ones with 360 obstacle avoidance, they would likely bypass each other.

 

[MAvic_South_Oz]

The above link is a video showing me trying to get my mavic 3 and evo lite+ to hit eachother,

Interesting test.
Both drones sensors would work well with larger items, good to see them pick up something drone size.
Would be good to know if at speed there'd be enough time to react, especially with 2 drones closing on each other in normal speed . . . sports mode ? Probably not going to help ?

We know the lateral coverage of DJI sensors is pretty good, what about vertical coverage ?
When you dropped one drone down and up from 0:37, did you attempt to fly straight in again and it stopped ?
Obviously vertically a couple of feet either way and there'd be close but safe passage for drones to pass each other anyway.

 

[Ty Pilot]

Why am I posting this video? It shows that even if the skies were cluttered with drones, especially ones with 360 obstacle avoidance, they would likely bypass each other. . . . . .

. . . . .as long as they're moving real slow and operating in ideal (for the OA) conditions perhaps. But I'll wager dollars to doughnuts that with those two drones moving at speed, head on, there will not be sufficient time for them to recognize/see each other and react.

The closure speed of two drones that can each do 40 MPH is doubled in a pure head on pass. 80 MPH is 117 feet per second. With an average detection distance of ten feet on each drone that would give the drones about 1/12th of a second to detect each other and stop.

 

[lookout]

. . . . .as long as they're moving real slow and operating in ideal (for the OA) conditions perhaps. But I'll wager dollars to doughnuts that with those two drones moving at speed, head on, there will not be sufficient time for them to recognize/see each other and react.

The closure speed of two drones that can each do 40 MPH is doubled in a pure head on pass. 80 MPH is 117 feet per second. With an average detection distance of ten feet on each drone that would give the drones about 1/12th of a second to detect each other and stop.

They're traveling and even the "momentum" of even these lightweight vehicles still don't stop immediately.

 

[Rchawks]

Thank you for the test. and for my money (really your money) I would leave it at that and not push you luck. there's alot of investment dollars there. Thank you for taking the risk though.

OMG PALM TREES!!!! I haven't seen those in 25 years.,,,,,,,,,,

 

[AlanL]

They're traveling and even the "momentum" of even these lightweight vehicles still don't stop immediately.

Amazon's drones weigh 80lbs. unloaded.

 

[old man mavic]

The above link is a video showing me trying to get my mavic 3 and evo lite+ to hit eachother, at least where they have obstacle sensors. I used the front ones for this experiment. This experiment wasn't a test of high speed approach, it was a low speed test. I wanted to see if the drones would see each other and stop. THEY STOPPED EACH TIME. Or, they tried going around each other. I could not get them to collide no matter how hard I tried.

I might make a higher speed collision risk video, but I don't know if im brave enough. But I expect they will avoid eachother or break. Breaking would be bad though. I would have to set them both to AVOID obstacles.

Why am I posting this video? It shows that even if the skies were cluttered with drones, especially ones with 360 obstacle avoidance, they would likely bypass each other.

well it was very nice of you to put your drones at some risk ,but in reality as others have said ,comparing your tests to a real life scenario, with to very fast and quite heavy drones closing on each other at the same altitude ,would i can almost guarantee end in disaster

 

[Don Testme]

well it was very nice of you to put your drones at some risk ,but in reality as others have said ,comparing your tests to a real life scenario, with to very fast and quite heavy drones closing on each other at the same altitude ,would i can almost guarantee end in disaster

What if I send them at eachother at 30mph at same altitude?

Maybe I should just leave it up to the commercial companies to test.

 

[Don Testme]

. . . . .as long as they're moving real slow and operating in ideal (for the OA) conditions perhaps. But I'll wager dollars to doughnuts that with those two drones moving at speed, head on, there will not be sufficient time for them to recognize/see each other and react.

The closure speed of two drones that can each do 40 MPH is doubled in a pure head on pass. 80 MPH is 117 feet per second. With an average detection distance of ten feet on each drone that would give the drones about 1/12th of a second to detect each other and stop.

Perhaps if commercial tech improves, things will change.

 

[Ty Pilot]

Perhaps if commercial tech improves, things will change.

Perhaps but as an example, look at Tesla's and their OA and self driving capabilities. I would say automobiles at that level have some very advanced tech, yet there have still been accidents and fatalities where those systems were in situations where you would think that they were sufficient but weren't.

 

[Don Testme]

Perhaps but as an example, look at Tesla's and their OA and self driving capabilities. I would say automobiles at that level have some very advanced tech, yet there have still been accidents and fatalities where those systems were in situations where you would think that they were sufficient but weren't.

Your right. The 1st tesla self driving car I ever heard of crashing was like over 10 years ago easily. It was driving by visual scanning only. I couldn't believe it? No lasers. No sonar, No RADAR!, Just camera data! that sounds insane to me.

What had happened was a semi-trailer truck was doing a u-turn on a commercial highway where people and trucks were allowed to do u-turns. The tesla car thought the trailer of the Semi Cab was a billboard. The AI simply thought it would go under the billboard (which wasn't a billboard) and SMASH !! i THINK THE OCCUPANT(S) DIED.

LESSON: SELF DRIVING CARS ARE NOT SELF DRIVING. You still have to watch the road.

My new Car has a fisheye lens dome camera in the center front grill. At 35mph and above it will keep my car in between the traffic lines (solid or spaced or reflector type). It will also slam the breaks if it sees my car closing in on another car too fast. When it slams the breaks, alarms go off everywhere. It scares you to death. Basically, making you hit the brakes before it has to do it for you.