This morning, Banner was banished from the kitchen, again. She doesn’t like having her freedom of choice impaired. At first, she stayed outside the perimeter being cleaned. (I don’t know how she knows that precise point, except that she is regularly asked to stay out of the kitchen when the floors are being cleaned.) Later she moved into the room with Emily, being very offended as she always is about the use of floor cleaning tools. (“Why ya gotta do that?”)
Banner has been neither curious nor amused. She does not see this gadget as a threat. Emily, however, interprets any shadow or almost imperceptible sound as a threat and goes into full-throttle assault. (on my ears) At 6, she is beginning to slip, missing a couple of deliveries. The gadget is fairly quiet and unnoticeable except for the sound when solution is sprayed, and that isn’t bad either.
I’ve been watching this little self-navigating device for a week now. It covers the area well, and avoids obstacles surprisingly well. It’s a bit like a blind rodent in that it strikes surfaces then backs up and strikes them again. Clearly it is a trial-and-error learner, but while watching, it isn’t always easy to recognize where it will go next as it executes its algorithm. It moves forward in a branching Y pattern, first to one side, then it backs up and goes to the other side. After it completes an “area” by its definition, it may run back to a baseboard, then follow the baseboard closely around a perimeter, or it may drive across an area and start some points again. I’ve watched it navigate obstacles tightly such as legs on chairs, the curved base on the legs of a stationary bike, rugs and the edge to a step-in shower. It also responds to drop-off edges surprisingly well. It cleaned the bath floor without falling into the shower, and yesterday I watched it clean the edge of the first step to the basement stairwell, backing and moving forward. I decided it would be safe and turned my back, only to find it cleaning the top of the wood, which is about 3/8″ higher than the floor, a little later. But it did not fall onto the next step.
Watching it is like watching a toddler learn to walk. Some actions are predicable, others are curious. I can’t tell whether it is learning about the shape of the available floor or doing all trial-and-error navigation. It appears to have some memory, but that may be a deceptive incidental observation, except that it will park in the original spot if it doesn’t run out of battery. In any case, there is some level of entertainment, if not studious relaxation, just from observing.
So far, it hasn’t found a place to be stuck, even in tight fits. Before starting the mop, I move some things that are on the floor, like the desk chair, a rug or the guitar stand, onto a rug so it can use the battery more efficiently (with less wasted effort to navigate). I leave a navigation path for other things, like the kitchen trash can. For this house, built about 3 years ago, we used all hard surfaces and selected furniture that doesn’t harbor dust bunnies. After about 40 years with Newfs, I hope to have learned a thing or two, but you never quit learning. The loft floor is engineered wood (hickory) and the basement floor and dog room floor are finished concrete. The main floor is otherwise tile. We chose a tile pattern that was very light in overall color, and it was modeled after finished concrete, so there are color variations. It has a deco tile about every 8-10 squares. A couple of times I found the gadget circling, as if it perceived a color spot on the tile, which was actually part of the tile. Yesterday I cleaned the living room and entry floors. It is good for almost 2 hours of battery use, but the cleaning fluid will run out in-between times. I start by wetting the mop pad (not soaking it), and yesterday I sprayed an area on the tile while it was cleaning rather than stopping it. I’m not sure whether it would pick up where it left off once it is turned off to refill, and I didn’t want to restart the entire floor’s navigation. After a little while, I conceded and refilled the cleaning solution, and it didn’t go back over the entry floor.
Those floors were fairly clean, so I recharged the battery and restarted it in the kitchen. The kitchen floor was dirty, a real test. (I used it first in the bedroom, which was also a good test.) This tile has a glazed, honed surface. A finished concrete floor was originally planned, but apparently those don’t hold up well to big dog toenails, and some dyes/stains are toxic. The integrity of the main floor surface was especially important, to allow the surface to be sanitized, so a glazed floor was imperative, as was traction. The honing on the glazed surface provides very good traction, even when the tile is wet. The weak link for tile is always the grout. We used “Power” grout, which includes a urethane hardener, with epoxy grout in the shower. Our tile resource was certain that this would hold up, but even the epoxy grout at the lab gets dirty and needs special cleaning (although it is more easily cleaned), so. . . I didn’t hold my breath. And it hasn’t. After yesterday’s cleaning, using the same pad as the living room for a first pass, this morning I used the tile brush on a broom handle to do a quick clean-up before restarting the mop with a fresh pad, and I cleaned a drop of dried egg yolk that it probably would not remove. It is doing a nice job. For other floors, a quick vacuuming or sweeping is sufficient prep to keep hair from accumulating and being pushed around. I started it inside a corner of the kitchen cabinets, and it completed an area about 8′ x 20′, then it restarted inside the cabinet area and followed the baseboard and an invisible line across the open area and went back and forth in this area about three times in what appeared to be a random hit and miss cleaning. Next it ran back and forth across this area as if it were inspecting its work and looking for missed spots. Again, that may be too much assumption. So far, it gets the job done, with good coverage. The limitation is that it needs to be refilled at about 45 mins. There isn’t a sensor that shuts down the device when it is out of cleaning fluid, and if the battery runs out before it determines that it is “finished” and reparks, you may find it anywhere, like under a bed. If it parks, you will hear a musical notification that its battery needs to be recharged.
I generally prefer reusable materials, and there is a mop refill option that can be washed. I’ll also add an extra battery, to keep it running for longer periods of time.
As with any new product, your satisfaction will depend largely on your initial expectations. My expectations were for a maintenance level of support. I wanted some assist but plan to continue the heavy duty cleaning every so often, followed by a rinse mopping. This device fits my expectations and needs for basic support. Most of our floors are open, so I’ll set a timer, stop it and refill when I’m here, or not worry about it if I’m not. One reason I have been a holdout on the robotic vacuums and mops is that I don’t want a WiFi connected device that uses a camera and measures every inch of every room. This is the basic Braava which doesn’t use WiFi navigation, and reviews looked like it might work. Another was reservation about performance. And the worst reason was that an automated device might run over an “accident” and distribute it throughout the house. (If you haven’t seen one of these Youtube videos, you may not want to.) Now I’m thinking about giving up and getting the larger one that runs using a WiFi on a schedule and self-parks. Newfs are neatniks, after all; if they are in dire straits and no one is home to help, they look for a rug, and this device avoids rugs well.
VISUAL: The Bot met the rug head-on, as it did the chair, then chose a way to navigate: