Where Neato's Get Stuck Preventing Cleaning

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

Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 22nd, 2013, 2:38 am

Neato's get stuck in the following conditions making floors unsuitable for Neato cleaning:

1. A tight cul de sac with narrow opening between furniture it can get into but not out, from limitations of the navigation software. (edit: takes a particular shape, very narrow exit; works well under chairs and cabinets for instance).

2. An overhead pinch on the front edge of the bumper, where it can wedge itself into some low openings too tightly, and cannot withdraw for insufficient traction.

3. An overhead pinch on the turret, where the bumper passes in and the turret gets wedged, similar to above.

4. Climbing a narrow quarter round molding, or round edge of a molding, more than 1/4" in diameter. In this situation the wheels have traction only on the molding curve and at too steep an angle for the slippery polished surface limiting traction. Again software fails to withdraw from the impossible climb, leaving Neato stuck. Wide sloping edges allow Neato to pass over higher threshold moldings forming a ramp with lower grade.

5. Climbing square edged material 1/2 to 1 inch high, such as chairs with bent wood runners instead of legs, and bases of some stands, for speakers etc., some tubular furniture. Neato may climb these instead of being deflected by the bumper, and get stuck on top of them with the edge corner pressing into the bottom with too much friction. The bumper will engage only material 1.5" off the floor.

6. Deep pile carpets and some carpeting too slick for adequate traction. It's just not big and heavy enough to plough through. This limitation applied to previous Roomba vacuums used for a decade.

7. Cloth items left on the floor which can get tangled up with Neato shoving it around, same as with manual house cleaning. Lightweight mats on hard floors can be too easy for Neato to shove around instead of going over them; it neads an edge fixed in place to climb. This can lead to all sorts of trapping clutter similar to the tight spaces problem mentioned above.

8. Some loose wires and cables creating situations similar to problem floor moldings. Even when not trapping Neato, it will pull wires around and possibly unplug things in the process. [edit: it goes over most straight wires well; coils of thick cables is harder]. It is not necessary to tack wiring up onto walls, however; they can be blocked off on the floor against walls various ways. Neato will not usually draw wires into the rotating brush because of the brush guard (especially with the beater brush lacking gripping bristles; absence of an adjacent counter rotating brush grabs less as well).

9. Not exactly getting stuck, but failing to stop: doughnut or circular floor plans created by rooms, often kitchens or halls, with two exits. Neato will not recognize entering areas previously cleaned when the size of the whole floor plan exceeds the twenty foot diameter of the square areas into which Neato divides up the floor. Crossing too many entire work areas to reach a previously done area, it loses track of the cleaning plan. It will stop when needing to recharge, then continue on indefinitely.
This problem is solved by a mag strip barrier or closed doors eliminating one of the entries, breaking the circle.
(some doubts exist about this limitation despite confirmation by the factory; discussed in two threads:
http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16912
and older http://www.robotreviews.com/chat/viewtopic.php?f=20&t=15819)

10. Large spaces with more than 15ft to any point the laser can see; Neato gets lost and shuts down. Can be fixed by having furniture placed to break up the open spaces with some landmarks. Neato is not equipped to clean large ballrooms and the like, more commercial venues; industrial robots in such places typically use some installed guidance system such as floor marks or imbedded wires.

11. [added] Long featureless hallways where distance along cannot be measured for lack of landmarks; similar to above except only one dimension breaks the 15ft limit. Thanks to Neato Robotics for pointing out this condition (see following posts). It is possible indentations for doorways along corridors can serve as detectable navigation points -- they didn't say; a few halls without doors over 30ft might then be the only cases.
[added] Just dark spots, black tape etc., at ten foot intervals on one side might serve as landmarks. Or both sides.

12. [added] As mentioned in factory materials, cloth hangings under beds etc. can block the laser after poking under. Suggests putting boxes underneath to hold up.

13. [added] Certain conditions atop stairs disrupting Neato's normal motion close to the edge, such as thick carpet in one case, can cause responses to the cliff sensor to be ineffective despite sensing, and Neato will terminate by falling over the edge, sort of inadvertently. This can be another example of insufficient sophistication in close quarter maneuvering, following too simplistic rules. Very few instances reported of this, however. Boundary strips solve the problem, but Neato advised they need to be placed a few inches back from the edge to be effective, from still more limitations on maneuvering. To the extent maneuvering is blocked, by impassable carpet say, there may be no software improvement possible. There is a minimal requirement of the unit being able to move for any function to be effective. It is still peculiar if the cliff sensors are tripped; at least you'd think it should terminate stopped in good order instead of falling.

Housecleaning is terminated when Neato gets stuck. [edit] With software up to 2.6 at least, one case can damage the unit or floors: getting pinched overhead with the wheels slipping on the floor; they don't turn off and the treads or the floor can get worn. A gap in the software, no telling when or if will be fixed. Apparently the bot thinks it is moving when the wheels turn and does not bother to check its Lidar. Yet the software is sophisticated on this when failing to climb some obstacles forward, backing off and trying angled attacks etc. So it is a bit peculiar. Unless this was fixed in 3.1 -- am not going to try it. The forward sophistication is definitely lacking when backing up. Lack of a rear bumper leaves blocking the wheels as the strike indication, and when the wheels slip, it is clueless -- despite its Lidar.


These conditions require raising furniture or isolating with mag strips, adding slopes to moldings or replacing them, or modifying the Neato bumper to deflect from a wider range of obstacles, both above and below the existing bumper. Raising furniture allows cleaning underneath, but especially some chairs do not have legs very suitable for adding height very aesthetically, and get moved around. Very large chairs, tables, chests, cabinets can be more suitable for adding to the leg bottoms.

There is a simple modification adding support to the bottom of the Neato helpful with some carpet problems. Other modifications to the bumper can prevent getting trapped, but will not clean under things. It seems the only reasonable solution, though, to floor moldings and some furniture which traps the Neato, short of preventing entry to entire areas with such pieces. Mag strips would seem inconvenient in a complex and changing arrangement. Modifications require materials, work, and some minimal skill. See "Furniture Handling Mods -- Summary" http://www.robotreviews.com/chat/viewtopic.php?f=20&t=17372,
(other mods in "Neato Mods for Utility" post, with links to threads on different sorts of problems.http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16475)
Last edited by glnc222 on October 26th, 2013, 12:26 pm, edited 10 times in total.
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby a4refillpad » March 22nd, 2013, 10:30 am

Good list but I think the cause of no. 9 is still contentious. I have a large floor plan with this "doughnut" scenario and I've not seen this with either the xv15, xv25 or the vr100. I believe there is more to it than a simple matter of what is mentioned. Maybe factors such as reflections (or lack of) or complicated furniture layouts is confusing the robot's internal map just enough to cause the effects seen. Bit of a mystery to me.
current bots: Clebo Arte, Vorwerk VR-100 (one broken one new), Neato XV-25, Neato XV-12, Roomba 564, Evo Mint 2, Roomba 770, Hauzen VC-RE70V, Samsung SR8980, Karcher RC3000, Robomow RL350 (lawnmower)
previous bots: Samsung SR8845, Neato XV-15
User avatar
a4refillpad
Robot Addict
 
Posts: 400
Joined: March 23rd, 2012, 7:49 pm
Location: UK

Re: Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 22nd, 2013, 12:28 pm

Doughnut problem: reported by the factory besides users. It appeared to me this could be a consequence of shortcuts taken in the programming to use the lidar, whose raw data is location relative to each room's walls instead of an architect's blueprint for the house. There's a mathematical problem in using radar like data to determine location, using landmarks and such. It is not clear it remembers the entirety of the map for every room.

When returning to base it backtracks the path it used to initially reach its position (instead of finding the shortest route), and when it enters a new area it pauses and spins as though taking a scan to remember the transition points. This could be done entirely in relative terms, relative to how it sees the shape of walls from that position only. It recognizes the area when returning to the same transition point where the same relative positions will be seen, but when entering from another direction cannot recognize the same place from different relative coordinates.

There appears to be no absolute coordinate system for the entire floor. It takes extra steps to create a map with absolute coordinates and I don't think they did it. I explained how such coordinates could be calculated in another old post, without remembering the entire shape of every room, just overall boundaries to simplify it. But we don't know exactly how they use the lidar.
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby neato » March 22nd, 2013, 2:23 pm

The Neato returns to the next probable base location by taking the shortest PROVEN route through the maps it has created while cleaning.

The Neato cannot resolve a "doughnut" shape area (that it cannot see across as a single space) because it appears to be an infinitely long room or hallway. It will also have navigation errors on very long, featureless hallways.
User avatar
neato
Neato Special Guest
 
Posts: 330
Joined: July 13th, 2010, 7:04 pm

Re: Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 22nd, 2013, 3:21 pm

Neato's clarifications are much appreciated. The hallway point is surprising because at least the side walls of the hallway are visible to laser even though the ends cannot be seen; it may appear not completely featureless but actually is in an important respect. The problem is that no landmark can be identified along the wall to measure distance moved down the hall compared to side to side. Another version of the 15ft range limitation.

So placing some feature along the wall big enough for Neato to see would help in this situation, a floor lamp, side table, umbrella stand, small bookcase or something.

At some point certain "dead reckoning" methods of navigation might improve performance; emulate traversing a cave in the dark feeling the wall. This relates to the absolute coordinate system need, in that a difficult passage could become workable when placed between known places. A separate procedure can traverse the passage: keep going until the known place reached. A ring hallway surrounding a room would still be a problem -- never ending (maybe dub the "ouroboros problem".
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby a4refillpad » March 22nd, 2013, 8:03 pm

Again, even though @neato has made this statement, I'm not seeing it myself. I clearly have a "doughnut" shape area which it cannot see across as a single space.
current bots: Clebo Arte, Vorwerk VR-100 (one broken one new), Neato XV-25, Neato XV-12, Roomba 564, Evo Mint 2, Roomba 770, Hauzen VC-RE70V, Samsung SR8980, Karcher RC3000, Robomow RL350 (lawnmower)
previous bots: Samsung SR8845, Neato XV-15
User avatar
a4refillpad
Robot Addict
 
Posts: 400
Joined: March 23rd, 2012, 7:49 pm
Location: UK

Re: Where Neato's Get Stuck Preventing Cleaning

Postby Wing Nut » March 22nd, 2013, 11:04 pm

At the end of its task, rather than take a direct route down an uncluttered hallway, my Neato prefers to return to base by deviating off into a cluttered room using the two doors that open onto the hallway. So it peels off through the first door, growls slowly across a thick mat, threading its way between couches, tables and chairs, and then meanders out the other second door and back into the hall only a short distance from where it first peeled off. I assume it just doesn't see (recalculate or save) the shortest 'proven' route?
#1 Neato XV-11 SW 2.0 Board Rev 113 / 4400mAh Lithium Power battery installed Jan 2018 / Pet & Allergy pack installed Sept 2013. In July 2014, received and repaired a faulty #2 Neato XV-11, running Lithium Power 4400mAh battery since July 2015
User avatar
Wing Nut
Robot Addict
 
Posts: 357
Joined: October 30th, 2010, 1:38 am
Location: Australia

Re: Where Neato's Get Stuck Preventing Cleaning

Postby a4refillpad » March 23rd, 2013, 5:34 am

Wing Nut wrote:So it peels off through the first door, growls slowly across a thick mat, threading its way between couches, tables and chairs, and then meanders out the other second door and back into the hall only a short distance from where it first peeled off. I assume it just doesn't see (recalculate or save) the shortest 'proven' route?


Hi you're yet another example proving that the neato can cope with the "doughnut" challenge. I found that the programming is really quite clever. if you close the door that would normally take on the route back it will eventually work out how to use the shortcut. it will repeatedly tap at the door at first for a few minutes though. Try it yourself its great.
So it looks like the neato is capable of using the shorter route but just chooses to use the longer "safer" route in preference to the shortcut.
current bots: Clebo Arte, Vorwerk VR-100 (one broken one new), Neato XV-25, Neato XV-12, Roomba 564, Evo Mint 2, Roomba 770, Hauzen VC-RE70V, Samsung SR8980, Karcher RC3000, Robomow RL350 (lawnmower)
previous bots: Samsung SR8845, Neato XV-15
User avatar
a4refillpad
Robot Addict
 
Posts: 400
Joined: March 23rd, 2012, 7:49 pm
Location: UK

Re: Where Neato's Get Stuck Preventing Cleaning

Postby invenio » March 23rd, 2013, 8:19 am

glnc222 wrote:Neato's get stuck in the following conditions making floors unsuitable for Neato cleaning:


Although not perfect, my neato actually handles many of the following quite well. Actually better than expected.

1. A tight cul de sac with narrow opening between furniture it can get into but not out, from limitations of the navigation software.


I have an area between two couches with a narrow opening (barely enought for it to go inside) and it's able to go in, clean the area, and then come out.

2. An overhead pinch on the front edge of the bumper, where it can wedge itself into some low openings too tightly, and cannot withdraw for insufficient traction.

3. An overhead pinch on the turret, where the bumper passes in and the turret gets wedged, similar to above.


This does happen to me as well on some couches which are almost exactly the same height as the body of the neato. I use magnetic strips to keep it from going under.

4. Climbing a narrow quarter round molding, or round edge of a molding, more than 1/4" in diameter. In this situation the wheels have traction only on the molding curve and at too steep an angle for the slippery polished surface limiting traction. Again software fails to withdraw from the impossible climb, leaving Neato stuck. Wide sloping edges allow Neato to pass over higher threshold moldings forming a ramp with lower grade.


I have transitions and molding larger than 1/4" and there does not seem to be a problem.

5. Climbing square edged material 1/2 to 1 inch high, such as chairs with bent wood runners instead of legs, and bases of some stands, for speakers etc., some tubular furniture. Neato may climb these instead of being deflected by the bumper, and get stuck on top of them with the edge corner pressing into the bottom with too much friction. The bumper will engage only material 1.5" off the floor.


I have over 10 different chair types through the house, not to mention different leg styles for furniture. It's never gotten stuck.

6. Deep pile carpets and some carpeting too slick for adequate traction. It's just not big and heavy enough to plough through. This limitation applied to previous Roomba vacuums used for a decade.


No problems with any rugs, mostly persian type of different thicknesses so far (with the exception of what I mention below).

7. Cloth items left on the floor which can get tangled up with Neato shoving it around, same as with manual house cleaning. Lightweight mats on hard floors can be too easy for Neato to shove around instead of going over them; it neads an edge fixed in place to climb. This can lead to all sorts of trapping clutter similar to the tight spaces problem mentioned above.


This does happen, but the truth is that even when I am manually vacuuming the rug gets bunched up and I usually have to stand on it while vacuuming manually in the opposite direction. I don't think one can expect a robot to do a better job than a standard upright vacuum does.

8. Some loose wires and cables creating situations similar to problem floor moldings. Even when not trapping Neato, it will pull wires around and possibly unplug things in the process. It is not necessary to tack wiring up onto walls, however; they can be blocked off on the floor against walls various ways. Neato will not usually draw wires into the rotating brush because of the brush guard (especially with the beater brush lacking gripping bristles; absence of an adjacent counter rotating brush grabs less as well).


My neato has cleaned over extension and power cords without getting stuck. The "front to back" wire guards under the brush seem to work well.

9. Not exactly getting stuck, but failing to stop: doughnut or circular floor plans created by rooms, often kitchens or halls, with two exits. Neato will not recognize entering areas previously cleaned when the size of the whole floor plan exceeds the twenty foot diameter of the square areas into which Neato divides up the floor. Crossing too many entire work areas to reach a previously done area, it loses track of the cleaning plan. It will stop when needing to recharge, then continue on indefinitely.
This problem is solved by a mag strip barrier or closed doors eliminating one of the entries, breaking the circle.
(some doubts exist about this limitation despite confirmation by the factory; discussed in two threads:
http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16912
and older http://www.robotreviews.com/chat/viewtopic.php?f=20&t=15819)


I have a relatively large (at least 1500 sq ft downstairs area) with a center stair case which makes the main living area a doughnut shape. It will go around and clean this area without problem and systematically. No endless loops or other issues.

My downstairs open areas are as large as 25x30 in areas and it has no issues systematically cleaning the floor plan. It never repeats the same area.

10. Large spaces with more than 15ft to any point the laser can see; Neato gets lost and shuts down. Can be fixed by having furniture placed to break up the open spaces with some landmarks. Neato is not equipped to clean large ballrooms and the like, more commercial venues; industrial robots in such places typically use some installed guidance system such as floor marks or imbedded wires.


I don't think I have any single point that doesn't have a piece of furniture within 15 of a spot so I can't comment on this. It does easily clean rooms which are at least twice as long as 15 ft however.

11. [added] Long featureless hallways where distance along cannot be measured for lack of landmarks; similar to above except only one dimension breaks the 15ft limit. Thanks to Neato Robotics for pointing out this condition (see following posts).


My upstairs hallway is close to 30 feet long. No problem cleaning it, as well as the rooms along it's path.

Housecleaning is terminated when Neato gets stuck.

These conditions require raising furniture or isolating with mag strips, adding slopes to moldings or replacing them, or modifying the Neato bumper to deflect from a wider range of obstacles, both above and below the existing bumper. Raising furniture allows cleaning underneath, but especially some chairs do not have legs very suitable for adding height very aesthetically, and get moved around. Very large chairs, tables, chests, cabinets can be more suitable for adding to the leg bottoms.

There is a simple modification adding support to the bottom of the Neato helpful with some carpet problems. Other modifications to the bumper can prevent getting trapped, but will not clean under things. It seems the only reasonable solution, though, to floor moldings and some furniture which traps the Neato, short of preventing entry to entire areas with such pieces. Mag strips would seem inconvenient in a complex and changing arrangement. Modifications require materials, work, and some minimal skill. A number are listed in "Neato Mods for Utility" post, with links to threads on different sorts of problems.http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16475



The above is a very through and fine list of "possible" issues. I don't know if they have been fixed with firmware updates, or perhaps I am just lucky. I am a new owner of a neato robot (XV-11), and frankly I was expecting a lot more problems. I usually will do a little prep of the house before sending it off but I find it works remarkably well. I do use barriers so it doesn't get stuck underneath my furniture which is almost exactly the height the neato's body, but that is the only adjustment I have needed to make for my 3,600 sq ft house which is filled with furniture on almost every wall surface, not to mention in the middle of the rooms as well.
Robot: Neato Robotics XV-11 (with Vic's HEPA filter mod)
User avatar
invenio
Robot Addict
 
Posts: 175
Joined: March 19th, 2013, 6:16 pm

Re: Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 23rd, 2013, 1:26 pm

It does take very particular circumstances to trap the Neato, but cases can be reproduced at will and have been reported by users here. Obviously rare or the machine could not be sold.

The tight space situation does depend exactly on the shape, as it navigates between furniture legs and under cabinets ok. Mere inches can make a difference, as I fixed one by moving a power strip under a cabinet. Effect of lifting the back in reverse is involved. A turn inside an opening almost as narrow as the vacuum and with a V shape side pinch may do it. Originally mentioned mostly with bathrooms.

It does go over most electrical cords without issue, but coils of thick coaxial cables have proved an issue, compared to a single line gone over. Just tucking cords against the corner of the wall and floor tends to keep Neato off them as it cannot touch that spot; the bottom is recessed under the bumper, and the brush won't touch.
Only certain types of moldings are a problem, for example a 3/8" metal edging of carpet, or straight sided 1/2" high wood thresholds. It climbs all sorts of things, but I have reproduced some which trap it.
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby JEfromCanada » March 23rd, 2013, 1:28 pm

Another item to add to the list...

Making it under a bed / piece of furniture with lots of initial clearance, and then getting stuck under the loose material hanging under the furniture (as it covers the turret, obscuring the view).
Current robots:
Roomba Discovery 4210 (R.I.P. - Freecycled)
Roomba 560 (retired)
Neato XV-11 SW version 3.1.17844, LDS version 2.6.15295, with Pet Brush and Vic's filter (sold)
Neato Botvac 80 SW version 1.1.97, LDS version 2.6.15295, Board SW version 22753
JEfromCanada
Robot Addict
 
Posts: 653
Joined: November 13th, 2007, 5:16 pm
Location: Canada

Re: Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 23rd, 2013, 1:41 pm

Neato mentions this in their material, suggest putting a box under the hangings.
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 23rd, 2013, 2:38 pm

My upstairs hallway is close to 30 feet long. No problem cleaning it, as well as the rooms along it's path.

Doorways in hallways, with indentations, might be sufficiently different "features" for the lidar to distinguish and use as navigation points; up to the factory to say. Halls are pretty common so the "featureless" kind must be more rare. Also a hall 30 ft long can still have ends visible at least from the midpoint. Only a longer one would have a "land of the lost" interval in the middle resembling rooms too large. Take Neato to a hotel and annoy the guests running in the hallway...
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby Wing Nut » March 23rd, 2013, 4:30 pm

invenio wrote:I have an area between two couches with a narrow opening (barely enought for it to go inside) and it's able to go in, clean the area, and then come out.

I have seen my Neato deal with two similar narrow gaps - one it handles fine while the other can confuse and trap it. It handles this gap with (looks clever, but hard to reach!) or without the docking station located in the recess - https://www.dropbox.com/s/togztsmfg6pt5r5/NeatoTightParkDesk_TimeLapse.mp4

But if I happen to leave a new ream of printer paper near a corner wall or waste paper bin, leaving a similar sized 3 sided gap as in the video, then it can get trapped with its rear end raised up, despite open floor ahead - the way it came in.

I suspect that in the video, the gap under the old desk there's 3 solid vertical surfaces for the LIDAR and bumper to 'see', but when one side isn't isn't high enough to be seen by the LIDAR, as in the ream of paper example, then only the bumper interacts with it and confusion can set in.
#1 Neato XV-11 SW 2.0 Board Rev 113 / 4400mAh Lithium Power battery installed Jan 2018 / Pet & Allergy pack installed Sept 2013. In July 2014, received and repaired a faulty #2 Neato XV-11, running Lithium Power 4400mAh battery since July 2015
User avatar
Wing Nut
Robot Addict
 
Posts: 357
Joined: October 30th, 2010, 1:38 am
Location: Australia

Re: Where Neato's Get Stuck Preventing Cleaning

Postby invenio » March 23rd, 2013, 5:36 pm

Wing Nut wrote:
invenio wrote:I have an area between two couches with a narrow opening (barely enought for it to go inside) and it's able to go in, clean the area, and then come out.

I have seen my Neato deal with two similar narrow gaps - one it handles fine while the other can confuse and trap it. It handles this gap with (looks clever, but hard to reach!) or without the docking station located in the recess - https://www.dropbox.com/s/togztsmfg6pt5r5/NeatoTightParkDesk_TimeLapse.mp4

But if I happen to leave a new ream of printer paper near a corner wall or waste paper bin, leaving a similar sized 3 sided gap as in the video, then it can get trapped with its rear end raised up, despite open floor ahead - the way it came in.

I suspect that in the video, the gap under the old desk there's 3 solid vertical surfaces for the LIDAR and bumper to 'see', but when one side isn't isn't high enough to be seen by the LIDAR, as in the ream of paper example, then only the bumper interacts with it and confusion can set in.


Nice video demonstrating this. The official word from neato is that the charging station should have 3 feet of clearance either side and out, I noticed you have it placed into that small area, is that causing any problems?
Robot: Neato Robotics XV-11 (with Vic's HEPA filter mod)
User avatar
invenio
Robot Addict
 
Posts: 175
Joined: March 19th, 2013, 6:16 pm

Re: Where Neato's Get Stuck Preventing Cleaning

Postby tginder2003 » March 23rd, 2013, 6:30 pm

Just wondering if you can help me. I bought a couple of throw rugs and found that the Neato does not travel well across them while vacuming. It seems like the wheels are spinning but the tires dont grab the material to keep it moving. Do they have different type of tires I can put on them to make the Neato travel across them easier
tginder2003
 
Posts: 1
Joined: March 23rd, 2013, 6:21 pm

Re: Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 23rd, 2013, 6:56 pm

couple of throw rugs and found that the Neato does not travel well across

see links on deep pile and slick carpets in "Neato Mods for Utility" http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16475 Easy and effective; video.

featureless hall fix
Neato should supply dark plastic covered reflective pattern material placards similar to used in the charger, transparent to IR, with a different pattern, for use creating features in problem halls. Come to think of it, it's just a dark band needed, as the walls reflect already. So a bit of black duck tape will do.
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby Wing Nut » March 23rd, 2013, 9:46 pm

invenio wrote:Nice video demonstrating this. The official word from neato is that the charging station should have 3 feet of clearance either side and out, I noticed you have it placed into that small area, is that causing any problems?

It didn't always dock in this narrow under-desk space, but it' was also difficult to read the LCD and press the buttons. I did it, and another tight park example, just to prove a point - that it could. Some new owners weren't even inclined to experiment with their preferred location because it was narrower than what Neato suggests (IMO, no harm in trying). With our house on two levels, we now have the charger up on the other floor and carry the Neato back and forth as required.

However, it still rolls into this gap during a normal cleaning run, and the charger now upstairs is still located in a gap that's less than the 3ft either side that Neato advises yet it very rarely fails to dock completely (probably for other reasons - navigation got compromised elsewhere and now low on power).
#1 Neato XV-11 SW 2.0 Board Rev 113 / 4400mAh Lithium Power battery installed Jan 2018 / Pet & Allergy pack installed Sept 2013. In July 2014, received and repaired a faulty #2 Neato XV-11, running Lithium Power 4400mAh battery since July 2015
User avatar
Wing Nut
Robot Addict
 
Posts: 357
Joined: October 30th, 2010, 1:38 am
Location: Australia

Re: Where Neato's Get Stuck Preventing Cleaning

Postby glnc222 » March 24th, 2013, 12:47 am

3ft clearance rule for base
This factory advice may not be for docking even though about base positioning. After all, Neato returns to its start position where it can see the base back in a niche, and backs in pretty straight instead of from any wide angles needing lots of room. The advice could be for making the dock visible across the room assisting return, but is that needed? So I wonder if it has more to do with being able to get an effective recognition of the room shape from the starting point, having a clear view of the whole. If scrunched against a wall it cannot see landmark shapes along that wall very well for navigation by triangulation. But as one filk song said, you'll never never know my friend.
glnc222
Robot Master
 
Posts: 4835
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: Where Neato's Get Stuck Preventing Cleaning

Postby Wing Nut » March 24th, 2013, 12:56 am

Yes, I agree it's probably more important to have a clear manoeuvring space directly in front of the dock, rather than the floors left clear up to the walls on either side.
#1 Neato XV-11 SW 2.0 Board Rev 113 / 4400mAh Lithium Power battery installed Jan 2018 / Pet & Allergy pack installed Sept 2013. In July 2014, received and repaired a faulty #2 Neato XV-11, running Lithium Power 4400mAh battery since July 2015
User avatar
Wing Nut
Robot Addict
 
Posts: 357
Joined: October 30th, 2010, 1:38 am
Location: Australia

Next

Return to Neato Robotics

Who is online

Users browsing this forum: anjosva, clone1008, iangough1000, jdong, johnnyh, leonicholson, ps0001, RockinRay, rotorwash, someoneclever and 473 guests