New Neato FW Spotted 3.0.17235

[ilpirata]

I just got my new Neato XV-21 from an ebay seller. It came with 3.1

SW 3.1.17844
LDS V2.6.15295
Board Rev 64
Board SW 17624

What about the battery duration? Does 3.1 improves that in comparison with 3.0?

[tmnCA]

I just got my new Neato XV-21 from an ebay seller. It came with 3.1

SW 3.1.17844
LDS V2.6.15295
Board Rev 64
Board SW 17624

Is it the case that the firmware is different for XV-11, 14 and 21?

My XV-14 from Costco arrived Dec 28 and has 3.0.17235.

[a4refillpad]

@tmnCa @golfingjoey had a youtube video of his xv14 with the new corner cleaning. (many thanks @golfingjoey!)
Can't wait as it looks great.

[craigbla]

I have an XV-14 with 3.1 and I do have the climbing problem with speaker stands that are about .75" high as well as an Ikea Poang chair plus another wood frame chair and it's footrest - all of them are about an inch high. The climbing is annoying because it can cause the Neato to get stuck plus it does an even worse job on the borders since it's up in the air at an angle. I realize they there is a need to be able to cross uneven room boundaries but I wish I could set an option that would reduce this behavior. I came home today to find it stuck on a chair again - arrgh!

I can remove the chairs but that's a nuisance and there is no way I would buy enough boundary and keep having to put it down and remove it.

I hope this behavior is fixed or an option gets added to limit it because it causes the Neato to lose it's simplicity.

Overall, I think it's a good product and a better vacuum than I expected - I let it run in the same room repeatedly until it stops picking up much dirt - I would never have the patience to do that with my regular vac even though it has better suction.

[glnc222]

climbing problem with speaker stands that are about .75" high

Not a software change issue, just compromises required even for carpet - hard floor transitions. See "tubular chrome chairs" http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16549&p=111623&hilit=tube#p111623

At some level individual pieces must be marked for Neato to recognize when to attack and when to avoid, but there are some improvements that might be practical (it's not going to exhibit human sensory ability or intelligence). It could be better at detecting when it is getting stuck and should abandon its attack. The inappropriate aggression is also needed to escape some tight spaces, so there is a trade off. But it's also been observed unusually dumb, given it's capacities, about tight spaces, not rotating the best and acting blindly etc., so maybe that can be improved as part of the whole problem.

Funny they respond to complaints about missing bits of corners, but not these problems causing complete failures. Corners may just be simple enough for their limited resources. Maybe if more people complained. Or it's a lot harder than first appears. Devil's in the details.

[craigbla]

I went to lunch and let it try to vacuum my family room again but it got stuck so it's 0 for 3 in that room.

Since there's 2 chairs that it climbs I will have to remove them from the room - which also means I can't let it run on the whole floor since it will climb them wherever I put them.

I think an option to restrict barrier height is reasonable otherwise this behavior makes the Neato annoying/impractical for me and I suspect that it wastes power and puts unnecessary wear and tear on the vacuum and the chairs.

[glnc222]

There is no physical way to restrict barrier height -- it has a bumper, period. To avoid something it has to be high enough to depress the bumper. Were the bumper lower, it would reject every height transition like a rug or floor molding as impassable. What you want requires an optical sensor for the low barriers, programmable, possible except there is no way to program it as you want -- every low setting would fail at something it should pass; things have to be marked, typically by making them high enough to trip the bumper, rather than mag strips -- more for overhead problems or whatever. Automating things not as simple as they seem when done unconsciously with human intelligence, or even animal sense organs.

[craigbla]

I thought that 3.1 increased the climbing because I read an earlier post that said they down leveled to 2.x after updating to 3.1 to stop the climbing. If the OP is wrong then I guess attaching a thin strip to lower the bumper maybe .25" might work. Maybe a more expensive model could have a bumper height adjustment.

[crowone]

I thought that 3.1 increased the climbing because I read an earlier post that said they down leveled to 2.x after updating to 3.1 to stop the climbing.

!! Climbing of the back site in the corner, most when using a plinth.

Video link

[glnc222]

If lowering the bumper only 1/4" would work, it would seem raising the only problem furniture pieces so little might be simpler, and avoid Neato navigation effects. All sorts of 1/4" material and furniture leg footings in hardware. A photo of the problem footings Neato can't handle would be interesting. I raised one floor stand 1/4" because of this climbing problem.

[crowone]

For some this works. silicon stops



Link.

[craigbla]

I can raise the speaker stands easily but not the chairs so I may try the silicon pad.

[glnc222]

Similar to the tubular chairs case. Just in case unfamiliar, 1.5" (2 nomimal) x 2ft precut boards in some stores ( Lowes in US) could be fastened to the chair bottoms with velcro, when screws or brads unsuitable; Other possibilities: bent wire like coathanger inserted into Neato bottom edge screw holes. Plate across the front on velcro strips up to the top edge (maximizing hold against bottom pressure with leverage) -- adjustable down extension. Plastic from CD jewel cases thinner than 1/4" wood, thinner wood in craft stores (basswood). Aluminum sheets in hardware. Plastic ordered from Amazon, ServoCity, etc. Fanciest addition a hinged flap up for normal, down for lower bottom; maybe just a piano hinge itself (pad with duck tape for contacts).

[Should start another thread on Climbing if continued as not software related. pics can be repeated by Img address reference without uploading. The opposite problem, overhead pinch trap, has it's own thread -- took a while to solve but works well, extending bumper up instead of down]

[added: 3/32 hitch pin clips are heavy wire which fits the bottom screw holes, into vertical tubes tight]

[UnknownPerson]

P1010387.JPG (69.77 KiB) Viewed 3993 times

Ohh.. I've got that chair. My Neato hates it and I have to have unsightly boundary marker stuff on the leg. I'd much rather Neato left the damn chair alone!

[craigbla]

I was able to mostly stop the climbing on the Poang by adding bumpers from Lowes:

Doesn't work for the other chair legs because Neato will spin and climb them from the side but it seems to work with this one.

[kamiraa]

I ended up having to surround the base of my speaker stands with boundary tape . . . which sucks.

I used double sided carpet tape to prevent the Neato from moving the boundary markers

Overall, it worked but looks unsightly. I may have to try this bumper method.

The Neato still ends up trying to turn around in area's it barley fit into in the first place, then gets stuck when it wedges itself in. The silly thing about this situation is it will attempt to unstick itself for a dozen attempts before throwing an error. Each time getting burried further and further, which strains the motor. I believe this is what caused my constant "left wheel stuck" error. It probably broke the epoxy holding the motor to the base. I've moved furniture to "NEATO PROOF" the house, but it's getting pretty nuts to make this thing work downstairs.

Upstairs I've figured out the "NEATO PROOFING", and hardly gets stuck doing 3 good size rooms, and a bathroom.

[glnc222]

Doubt the wheel drive has enough torque to break it's mounting. On many surfaces when Neato is blocked from moving, the wheels will spin slipping. When they don't slip, the computer can detect the wheel is not turning and report stuck. This has been observed with some cable tangles. Depends on going forward or back, with different pressure on the wheels. It slips easily on carpets and slicker hard surfaces. The motors are protected with sensors, probably won't be damaged by anything even if not cleaning properly.

See the other thread: http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16736

[kamiraa]

Well got my new replacement Neato in yesterday. It arrived with the 3.1 firmware straight from Neato!!!

I picked the option to set a new battery, and allowed it to charge overnight.

Results???

52 minutes out the box first time, excellent considering I've always struggled with ~38-42 minutes even with a new battery replacement!!

The way it handled the first floor was very interesting. Before it would attack a single room (sometimes repeating areas), get stuck sometimes, and then move on into another room.

This one was all over the place tackling the perimeter, Within the first run out it had been in 5 of the areas, clearing the perimeter and even partially finishing out the rooms.

On the second charge it went back to work hard, and knocked out in the inner surfaces, moving from room to room.

It's back at it's base to get ready for it's 3rd run. I have full confidence it will finish the run before the battery is even up. THIS NEVER HAPPENS, it always has to give up before completely finishing even with 3 attempts. This extra run time is helping for sure.

Even though the floor was recently completed by my other Neato VX-21 (moved from upstairs) the bin was filled with dust and hair. With the new feature to get closer to the walls, those areas were never getting fully cleaned. Now it knocks it out.

I'm not sure if they received an even better battery set, new hardware, or the firmware is affecting the time but very impressive!

The best thing was at least this run out, it didn't get stuck in places where it got stuck before.

Go NEATO, great that this thing is becoming better and better with ever upgrade. I'm hoping this one I will be able to leave un attended in the future and NOT get stuck.

[a4refillpad]

disappointingly it appears that neato are only providing 3.1 for all new devices pre-installed. Still 2.6 on their website and no update as to when it will appear. As discussed in the forum the firmware revision on the new neatos are exactly the same as the version on my vr100 so it does not look like there's been any update on the software since it was released over half a year ago on the vr100.

[justflie]

Agreed. I thought our friend Neato promised this update sometime before Christmas. I remember him saying there are a few last-minute issues but they were working on them. Obviously the new firmware has not been posted for the XV-11. Disappointing, but hopefully comes out soon.