VX-21 blocked by area rugs?

News and information about the Neato XV-11 Robotic Vacuum. All discussion and troubleshooting questions go here.

VX-21 blocked by area rugs?

Postby dzmiller » November 27th, 2013, 4:33 pm

I have a new vx-21. I'm surprised how much it struggles to get up on area rugs from hardwood floors. In fact, it often quits trying to return to base and complains it needs its path cleared. Anyone else have this problem?
dzmiller
Robot Groupie
 
Posts: 64
Joined: November 27th, 2013, 4:29 pm

Re: VX-21 blocked by area rugs?

Postby glnc222 » November 27th, 2013, 5:32 pm

The only way to improve that for the moment would be with a carpet "slider" mod if that helps (simple cardboard), but also might require increasing traction of the wheels -- which requires crafting replacements for them. See "Furniture Handling Mods" http://www.robotreviews.com/chat/viewtopic.php?f=20&t=17372&p=118943&hilit=slick+carpet#p118943

Depends on whether the problem is wheels slipping on the edge of the carpet or drag from the carpet on the front.
Generally to climb obstacles a stiff material must thrust up the brush guard to lift the front of the Neato, but thick soft material might not work right. Details of the particular rug would be needed to analyze. Maybe something new is needed. Whether the Neato stops with wheels spinning against the edge of the rug or when the wheels are on the rug is important.

Contact privately if interested in spending on enhanced replacement wheels instead of doing it yourself.

[note your other post was entered twice somehow; you can go in and remove the duplicate if you like]
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby dzmiller » November 28th, 2013, 9:53 am

thanks for the reply. The unit has trouble getting it's wheels up on the rug. As the front of the unit contacts the rug the front wheels lift and lose traction on the hardwood floor. Roomba also has some trouble with these rugs, but always returns to base. Roomba (630) exhibits more advanced programed behavior to solve these types of issues, IMO.
dzmiller
Robot Groupie
 
Posts: 64
Joined: November 27th, 2013, 4:29 pm

Re: VX-21 blocked by area rugs?

Postby glnc222 » November 28th, 2013, 8:56 pm

You can test the value of enhanced wheels with the rubber band method, doing a spot clean to see effect on travel. A slider is probably needed as well for the carpet, trivial to add. The band won't stay on permanently but may work for a brief test. Position a foot away from the rug edge with wall sensor towards the rug, and Neato will turn right going onto the rug.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby third_deg » November 28th, 2013, 11:06 pm

Neato copied Roomba R2 suspension design. The pivots or in the rear. A forward pivot (As with all 5XX and up Roombas) is dynamically superior for forward traction. This is why iRobot made the change.

Trilobite wheels move straight up and down which is also superior.

In short a rear pivot reduced traction.
third_deg
Robot Master
 
Posts: 507
Joined: November 12th, 2013, 2:01 pm

Re: VX-21 blocked by area rugs?

Postby glnc222 » November 28th, 2013, 11:36 pm

Excellent observation, though I'm not quite sure of the physics. Interesting iRobot changed the mechanics, as they all look similar.

The Neato pivot lets the body rise over an obstacle while wheels are pressed to the floor by the spring, and weight is taken up by the obstacle ([edit] and to the rear from tilt), reducing wheel pressure on the floor a lot, reducing traction (along with reduced contact area in some cases, as well as slippery surface area of the obstacle or rug). I am not sure it is the pivot or suspension type as much as the front end making a difference in weight on the wheels, and some drag effects.

Both need wheels on the center line for turning in place, however suspended. The circular Roomba's front may transfer less weight to the obstacle than Neato's D-shape. Roomba also has a caster in front? and what effect does that have -- maybe reducing forward drag on the obstacle? If I had a Roomba I'd measure all this, but don't.
Anyway, if you want to use a Neato, the only option is enhancing the tires.
Much appreciate your comment.

[edit] An interesting thing about the Neato is some simple and inexpensive modifications by users can improve the handling of problems presented by furniture and increase its range of operation somewhat (see Furniture Handling Mods http://www.robotreviews.com/chat/viewtopic.php?f=20&t=17372). The other designs don't appear to me to have enough greater range as made to offer the same potential, and have no fixes available. There are inherent trade-offs in the designs, as the Electrolux had the bigger wheels needed but could not fit under as much low furniture. A much more expensive, industrial scale bot, could be made today and do everything but not be suitable to consumer budgets. So it's all a matter of taste and personal preference at this stage of technology.

What happened in this industry is the cost of digital electronics for controlling bots fell to a point these devices were practical (over a decade ago), but the limited power supply from batteries is still way below what would be needed to compare with a regular vacuum, at least for cleaning carpets. So it's a new sort of cleaning operation with its own pros and cons up to consumer taste, and especially the furnishings to handle. The greatest advantage seems to be on hard floors where a full power vacuum was never needed in the first place. So then the occasional area rug presents a conundrum.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby glnc222 » December 3rd, 2013, 11:59 pm

Further suspension detail:
Neato not only has a wide front going onto the rug compared to Roomba's curved front, Neato's front is the cleaning head kept pressed to the floor. So the front is buried into a thick rug with greatest drag compared to Roomba's caster keeping the body elevated, with the cleaning head farther back. These bots things sink down into plush carpets. The slider mod may improve Neato's attack on some carpet edges.

Roomba, if I recall correctly, also has a separate moving suspension of the main brush center head, so that it can be elevated some without pressing the body into the carpet as much.
Maybe some more complicated suspension of Neato's front would allow it to flex up over a rug edge while keeping more weight on the wheels.

I suspect such details are a little beyond the scope of what the makers do. They probably assume the bots don't work in every environment and stick to marketing for where they do work. I would think expanding the range and satisfied customers would be wiser in the long term, but I don't answer to the venture capitalists. Fortunately some improvement is possible to the Neato with a few adjustments by the user. If they just supplied a couple of alternative parts, changing nothing on the main product, would accomplish much -- and shouldn't cost much, either.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby third_deg » December 4th, 2013, 10:48 am

With the suspension pivot ahead of the wheel, motor torque presses the wheel harder into the carpet. The opposite is true when reversed.

So the Neato has more traction capability backing up than going forward.

Also, it's important that the bumper covers all high points. Neato laser turret is a liability.

Neato goes straight down every hallway. So when it encounters the rug (or even a high transition as in my house) both wheels encounter the obstacle at the same time, forcing the mobility system to lift the entire robot weight at once. If it encounters the obstacle at an angle, it can climb with 1/2 the effort.

In my house not only would it not clear a transition, but it refused to try another way. Ultimately giving up.

Finally, they don't suspend the cleaning head. (However brush rotation is in direction of travel).
third_deg
Robot Master
 
Posts: 507
Joined: November 12th, 2013, 2:01 pm

Re: VX-21 blocked by area rugs?

Postby glnc222 » December 4th, 2013, 1:24 pm

Also, it's important that the bumper covers all high points. Neato laser turret is a liability.
Easiest fix of all with a wire loop on the front (Furniture Handling Mods); best done at the front detecting while there is more room to maneuver. Surpised laser guidance unaffected by thin obstruction.

If it encounters the obstacle at an angle, it can climb with 1/2 the effort.

3.1 software at least has a maneuver I get anyway, to back off obstacles stopping it and repeat attacks at angles, succeeding on retries when failing straight on.

In my house not only would it not clear a transition, but it refused to try another way. Ultimately giving up.
There may be a difference from whether wheels spin slipping at the obstacle or are stopped, which it seems to detect. The backwards traction difference gets involved because it seems to check the laser going forward but not backwards, getting stuck going backwards if it fails to detect wheel blockage -- used as substitute for the missing rear bumper. Loose ends in software.
Increasing the tire traction helps with obstacles where the wheels can be observed spinning on the surface.

The front brush position requires the rear pivot else wheels would be too far back; maybe vertical suspension possible. [edit] also a front pivot would make forward torque raise the front, inconsistent with front intake to be pushed down, the way the existing rear pivot makes that high back flip -- which needs some limiting for its own problems (simple thread loop).

they don't suspend the cleaning head.

With the head at the front on end of a see-saw pivoting on the center wheels, it adjusts to floor without separate suspension, at least for that aspect. Maybe the center positioning demands the separate suspension more. Be nice to have simulator software for it. Let the AI design it.

Go back to Electrolux design, stop cramping everything. Uprights don't fit under things anyway. Put a telescoping food service tray on top.
Last edited by glnc222 on December 6th, 2013, 11:31 pm, edited 1 time in total.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby third_deg » December 4th, 2013, 4:37 pm

OK. Those are great software fixes. I tested an x-11.

As to the see saw front. A suspended cleaning head only has to suspend the weight if that part. It's harder to balance the seesaw as you are positioning bins, and fans, etc...

So the suspended cleaning head generally gives you the greatest control of mobility impact.

Any weight not on the wheels is traction lost (subtracting the minimum needed for cleaning efficacy)
third_deg
Robot Master
 
Posts: 507
Joined: November 12th, 2013, 2:01 pm

Re: VX-21 blocked by area rugs?

Postby glnc222 » December 6th, 2013, 4:41 pm

A curious possibility occurs to get the best of both worlds in traction: a two wheel belt track with one wheel raised, for single wheel motion when not needed. When falling into deep carpet push down the raised wheel and switch to tank track mode. Requires more sophisticated operating software. The hope is to fit into the low slung units which can fit under furniture, and still get the strength to handle the problem carpets.

Still the improvement from just increasing Neato traction with a softer, occasionally replaced tread should be explored first. Traction was found increased fifty per cent this way. All it needs is a wheel with raised rims to hold in a 1/4 inch of tread material, instead of screws in the crude model, for quick attachment. Can be crafted from thin disks or molded and 3D printed entire.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby third_deg » December 6th, 2013, 6:13 pm

Can you boost spring tension on the Neato any higher? Trilobites have just a bit less spring force than robot weight, and a CG centered over the wheels, and large number of castors.

Really performs well.
third_deg
Robot Master
 
Posts: 507
Joined: November 12th, 2013, 2:01 pm

Re: VX-21 blocked by area rugs?

Postby glnc222 » December 6th, 2013, 6:55 pm

I mentioned that in one of the ruminations on suspension back aways, and it seems possible. Increase the spring tension and put on some steel plates (lead being toxic) to balance. Difficult home made needing particular spring etc., compared to crafting wheels, and unknown value till measured and tested. Maybe even a motor adjusting spring tension is possible. If making these things I would get a computer simulation set up to analyze different designs, so many variables. Maybe more intuitive to mechanical engineers.

It's a whole different scale of forces compared to furniture carts or what not, and needs new thinking, more like model airplane makers. It took a while to realize heavy duty plastic was not needed for the carpet slider mod and cardboard adequate. Vorwerk made a mistake initially supplying some very hard rubber treads different from Neato's with lots of complaints, now corrected (but oh so durable, superior Euro quality). For all that extra price? I don't think they fully understand what they are making. Or that it does not replace full size vacuums leads to a view they don't need to.
New entries are showing up and it could get more interesting. Keep watching the skies (for that Amazon drone delivering a vacuum). How about a drone you hire to drop down and mow the lawn?
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby glnc222 » December 7th, 2013, 1:57 am

Noticed brand Dave Brown Flite Lite model airplane wheels have a foam tread of the sort sought for improving Neato, and even a 3" size offered -- but a little too wide maybe at 3/4", and the hub not the same, could be as much work as making the whole wheel from pipe fitting I used. But shows the factory could get this sort of wheel tread supplied if they wanted to improve the product.

All sorts of robotic wheels available for experimenters, but not for Neato's oblong shaft -- might be adaptable, unknown. Stuff for making small robots has become quite an industry. Interest probably grows from modeling hobbyists, the way home made computers came out of Amateur Radio.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby frenchie » December 9th, 2013, 8:57 am

So, there seems to be 2 possibilities to handle high area rugs :
- slider under the robot (see forum thread about this)
- move center of gravity towards the front of the bot ? (stick flat lead weights on the bumper ? what weight is enough ? they make such weights for model RCs and boats to adjust the weight distribution...)
frenchie
 
Posts: 34
Joined: June 28th, 2013, 8:35 am
Location: CT

Re: VX-21 blocked by area rugs?

Postby third_deg » December 9th, 2013, 9:43 am

Why add weight to front rather than over wheels?
third_deg
Robot Master
 
Posts: 507
Joined: November 12th, 2013, 2:01 pm

Re: VX-21 blocked by area rugs?

Postby frenchie » December 9th, 2013, 10:02 am

To change the weight distribution so that the front flips down sooner when going over a transition or a high pile carpet...
I'll try the cardboard first, it's a proven solution
frenchie
 
Posts: 34
Joined: June 28th, 2013, 8:35 am
Location: CT

Re: VX-21 blocked by area rugs?

Postby glnc222 » December 13th, 2013, 4:42 pm

problem rug mysteries
Neato handles some height transitions and area rugs but not others, and there is little information on just what types are a problem. Any photos to show composition and measurements of thickness could alert customers to where Neato can run and whether any adjustments would help or not. One way to measure a rug is place a book on the edge for a point readable on a ruler.

It has been known since earliest Roomba's certain deep or fluffy carpets, shag rugs, fringes or what not, cannot be traversed by these bots with small wheels and low ground clearances -- just like needs for off-road vehicles. Some fluffy bath rugs it can climb but move on like through molasses. It would be better if it would not climb them...
So the environment has to be analyzed for using these products.
glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: VX-21 blocked by area rugs?

Postby third_deg » December 13th, 2013, 7:23 pm

Both trilobite and Roomba 5XX and up have low ground clearance but use a floatation based system with larger diameter wheels, spring pressure and CG location to traverse obstacles.

Later model Roombas can traverse some objects as tall as an inch, as can Trilobite.

Mobility is an area the Neato is a bit further behind in. I'm sure no company wants to list the liabilities though.
third_deg
Robot Master
 
Posts: 507
Joined: November 12th, 2013, 2:01 pm

Re: VX-21 blocked by area rugs?

Postby glnc222 » December 14th, 2013, 6:20 pm

glnc222
Robot Master
 
Posts: 4541
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Next

Return to Neato Robotics

Who is online

Users browsing this forum: calebleemcd, moradi2, vic7767 and 357 guests

cron