Spence’s methodology brilliantly exposes the flaw in judging autonomous systems by extreme stress tests rather than their intended environmental harmony. It is a sophisticated pivot from superficial benchmarking to a genuine understanding of contextual intelligence.
Deep Dive
Prerequisite Knowledge
- No data available.
Where to go next
- No data available.
Deep Dive
Navimow X4 Robotic Lawnmower | The Long Way Around
Added:[music] >> Normally, the review process is pretty straightforward.
You get a product, you unbox it, run through a few tests, formulate an opinion, and then you make a video about it. That video would be typically 10 minutes, 15 minutes. This video is considerably longer than that.
And the reason is simple. The conclusion that I reached at the end of this video is nowhere close to the conclusion that I started with.
I started with an observation. That observation led to a hypothesis.
I tested the hypothesis. Some of the results strengthened it.
Others didn't. As a matter of fact, some of those results actually challenged it.
Which led to additional testing, new observations, and a different conclusion ultimately than the one I expected to reach. What you're about to watch is that process unfold in real time.
You're going to hear observations, assumptions, and conclusions that I genuinely believed in the moment. Some of them held up. Others changed as more information became available through testing. What we have here is a Navimow robot lawn mower by Segway. And if you're looking for app features, a long list rundown of what this can do, walk-throughs of additional features, this is not the video for you. This is the equivalent of me going to test drive a truck, a $120,000 say Ford Raptor or something. I'm not interested in the air conditioning and the seat heaters and all the bells and whistles. I'm taking that thing out to the nearest set of sand dunes, and I am ripping on it. And I'm just seeing how it does performance-wise. So, with that said, we're not going to evaluate this robot lawn mower, a very sexy robot lawn mower, by talking about the user interface for 10 minutes. I wanted to know what happens when it faces terrain, obstacles, difficult conditions, extreme mowing angles, crazy things that you're supposed to cut other than like residential grass. Basically, the real world environment out here in the mountains of Colorado. So, before we jump in, I built a little montage to show you exactly the environments that this machine would face out here in Colorado. Before putting any mower to work, I wanted a repeatable way to evaluate performance. Over time, I've developed five dedicated test areas [music] around my property, each designed to challenge a different aspect of a mower's capabilities. The first area is the jungle. This section is mostly flat, but the grass grows thick, dense, [music] and overgrown very quickly. The jungle is designed to test overall cutting performance and determine how well a mower handles heavy vegetation. Next is the hill climb.
[music] This area features steep grades, uneven terrain, and mountain grasses.
The hill climb is designed to evaluate traction, stability, and overall terrain capability when conditions become more demanding. The alley is a narrow section with rough ground, a mix of grasses, weeds, and limited room to maneuver.
This area helps evaluate navigation, obstacle handling, and how [music] effectively a mower can move through tight spaces and inconsistent terrain.
[music] The leech field provides a more open environment with mountain grass, weeds, pine cones, and natural debris.
This area is useful for evaluating mapping, performance, navigation, and overall cutting consistency across a variety of conditions found on larger properties.
Finally, there's the gauntlet. This is the most demanding area on the property, >> [music] >> combining steep climbs, off-camber terrain, obstacles, rough ground, and thick vegetation into a single challenge. Many mowers simply get stuck here, making it an excellent test of traction, power, cutting performance, and overall capability. [music] Man, at this point, I really did not know what to expect. The Navimow looks refined. It looks well-engineered. I referred to it several times as if Knight Rider were a lawnmower. But, appearance and capability aren't always the same thing.
The first step is simple.
Put it in difficult terrain and see what happens. All right, guys.
We are in the jungle. You can see it behind me.
And this is the Navimow by Segway. And it is so sleek and so cool-looking and so like race car Knight Rider F1 car-looking.
I almost don't want to do this test. I almost don't want to get it dirty. It's uh it's very, very nice to look at. Super cool features. The real question is, how's it going to do on the stress test?
And I do think, if I recall, the folks at Navimow specifically used the word stress test. All right, this is our interface, our user interface. Really simple. We've got front and back.
And look at these wheels, like I told you.
All the way, almost like a zero turn.
Pretty cool.
But, the real question is, will it cut?
All right, buddy. Come on over here.
Let's talk about it real quick.
So, this was our first uh pass.
And it makes me wonder like, is there something I'm missing here? Did I Did I Is there a blade that I haven't like got out of the box yet? I don't know. Let me Let me open not open, but let me I'm going to hit stop.
I'm going to lift the mower up.
Right? Just want to be careful not to do this while I'm at Yeah.
There are the blades.
You can see them right there.
So, we've got the blades.
And as this spins, blades are coming out.
Okay, we're good. Let's give it another go. What I'm seeing here is in looking at it, this is some pretty tough terrain.
This is really thick terrain for this machine. But, let's give it some half passes to see if we can get some cutting done.
What I have immediately noticed as this weather moves in here, we've got snow more snow coming in May and some rain over the next few days. So, few more tests try to get them done.
But, with the jungle, maybe more geared to classic residential lawn. Different use case. That's all. Different use case. For something that thick and I would imagine even the folks at Navi Mower Segway would say, "Hey Phil, it's not meant for that." But, that's the purpose of a stress test.
Let's put it through that and see what the limit is. Speaking of which, we've got the hill climb next. Let's give it a go.
Let's get down to the start of the hill climb here.
Again, uh the power of this thing is not in question.
All right, so we're down at the base of the hill. And in fairness, like I said, the mower might not be geared for this, no pun intended. It's also it sits very low to the ground. So, I'm not going to intentionally try to run it up over like these big mounds, cuz it's going to high center and it's not going to go. So, let's give it a chance and let's just run no blades and let's see how it does up the hill.
Just the the motor itself and the traction it has. Here we go.
That's pretty freaking impressive. Come on.
Come on.
Get up.
Oh, wow.
Wow.
The issue it's facing right now, you can see right now, it's on the ground. This is a very low machine and it's got a bumper that's so low to the ground. So, stuff like this, it's going to struggle in, but this is the stress test. So, let's turn it and get it downhill.
We'll spin the blades and give it another go. I am actually, given what this machine looks like as far as how low it is to the ground, I'm very impressed that it got up the hill like this.
Those blades are so quiet.
From this distance, I can barely hear them.
Let's get up the hill with the blades on.
Come on.
Come on.
Uh back out of that.
Take this route here.
Little to the left.
Come on.
You can do it.
To the right.
Props to this machine.
Look at that. Look at that.
Yes.
Wow.
All right, we've got the rain moving in.
So, all we got left is the alleyway here where we're going to test the suspension articulation, the hill climb.
Impressive as far as the ability of the machine, the the torque and the power to get up a hill, especially how low it is to the ground.
Uh it cuts it, but it would take many, many passes to get it down to an acceptable level, I think.
So, let's set the camera up on this section, the alleyway, right? And we'll run it front to back, see how it does cutting, and also how it handles all the little dirt potholes and things like that. Here we go. Let's spin it up and see what happens.
Will it get to the other side?
Blades are on.
Get down into this grass here.
Blades are still cutting.
Ah, we're stuck.
Back out.
Take it up to the right.
Ah.
Come on. Come on.
There we go. There we go.
Come on. You got it.
All right, we got the machine behind us.
I don't want to get close enough to the mic with the wind picking up, but as expected, the cutting ability of the machine not really there, but I was impressed except for the one time I got stuck, the ability of it to traverse uneven terrain. So, what's the takeaway from all three tests inside the fence line?
Uh it's just not meant for this use case, in my opinion, based on what I've seen.
The blades are only about this big, and they're spinning pretty quickly, but they're not meant to tackle that really heavy, thick mountain grass. And I think, like I said, it says it even in the instructions, right? Things taller than 4 in. But, thought to put it through the paces.
So, the initial test gave me some answers, but they also raised a few questions. The Airbnb offers a completely different environment, different terrain, different grass, and hopefully a better understanding of what this machine is capable of. Right now, in the moment, I'm still trying to figure out what type of machine I'm actually dealing with. All right, we're at the second property, still a mountain property, but the grass, I feel, is a better fit for the Navimo, and we're going to give it that chance.
All right, so we've done part of the front area with the Navimo and I got to say it's done better than it did at the mountain house, the true mountain house.
The grass is a little thinner here. It is able to cut it, but it takes a lot of effort. By that I mean, you got to cut it at 4 in then drop to 2 in and you got to go a little slower. So, probably not the optimal environment for this machine, but it is doing better.
>> So, by now couple tests under my belt, I'm really starting to think I understand the Navimo. I have some ideas about strength, I have some ideas about limitations. The next step is to figure out whether those ideas can survive additional testing. All right, we're back. We got better weather. It's a different day and I realized in the time that it took for the weather to clear that I have not or was not going to give the Navimo the chance that it deserves. We watched in the yard on the top of the yard, the hill climb, and on the bottom in the alley, it really struggled to cut. But so what, right? So what? We deserve The Navimo deserves to get the rest of the torture test out here on the leech field, out in the gauntlet at the end.
All right, maybe the blades aren't made to cut super big heavy stuff.
But by going through the rest of the test, which is something I was not going to do, but I need to do, it will also expose the other good qualities of this machine.
And let's just go over this this stuff real slowly here.
Let's push it over through here.
Oh, here we go.
So, this is good to see. We've actually got a pine cone stuck in the blades, which is keeping it from rotating.
All right.
So, good to know it looks like this deck, yeah, it does.
This is what happens, it moves up and down via electronic motor. Pretty cool.
But that little bitty pine cone in there caused the blades to stop. All right, let's try it again.
All right, 1.75 in. Let's get the blades spinning.
All right, we're cutting.
We are definitely [music] cutting.
All right, I am so glad that I continued this test [music] because this mower is proving to me that it can This is at a 1.75 [music] in. It can cut a pretty decent path through this lighter version of some mountain grass. [music] Okay, what did we learn in the Leech Field test that the deck of the mower truly can operate at a lower level, where I assumed in the much higher stuff it could not. It's part of the test and why we do it and why we're giving the Navimow and the engineers of the Navimow, if you're watching, a little redemption in this phase. Now, the grass here wasn't super gnarly, right? Like it was over here in the jungle, but it was able to handle a lot of it and even some of the lighter twigs. Uh I think it has potential. We're going to now take it down to the gauntlet and we're going to see what's what. It's some pretty steep terrain going up here. The cameras don't give you the best view or the best elevation change sort of concept, but it is If you had to sprint up and down this thing like 10 times, you you'd be out of gas. So we're testing the I have no doubt that the Navimow will do well.
Look, I almost just fell the elevation change. I have no doubt that it's going to do well crawling up this with its all-wheel drive. No doubt. The real test is in some of this grass and some of the grasses here, will it still be able to cut? I'm going to lower the mower deck.
Well, I'm not going to run the mower over rocks um like these right here or those right there. And spin the blades.
If we don't get through this, we'll go back up to 4 in. I All right, not bad. Pause it right there, you can see we did get some cut going on up through here. Not a complete total cut, but it is cutting.
So far, no real traction issues.
Blades are still spinning.
I know, it's hard to hear them.
All right, let's go left here.
So, this is one of the key issues.
And this isn't the deck. This is See how low the bumper is?
It's obviously going to get caught. This mower was not designed to climb up a mountain or up a hillside in front of Phil's property, but it's good that we're testing it. One thing that really it has going for it, I'm telling you, is it is crawling pretty slow, but the the maneuverability that it has. Look at this, look at this, look at this.
These are totally independent of each other.
Right? So, you look at this weird configuration, you think, "Oh, man, it broke." No. It's It's like when it faces an obstacle like stick that doesn't allow it to turn, it's still it's still operational. It's still going to perform. As soon as I back this thing up, it's going to go right back to where it needs to be. Check these wheels out.
Isn't that cool?
I'm getting distracted.
Come on. Push, push, push.
I'm coaching it like it's a college kid.
Here we go. Back up.
I don't have the mower on. I'm just trying to get it up this area.
Until we're spinning here.
Spinning pretty good.
All right. All right. We made it up. We made it up that Let's get Let's go forward and right.
You got to get over that stick. You can do it.
Ah, come on.
Come on.
Told you I'd get it up. Come on.
It's going to go. Nice.
Come on.
This point, it's just about finishing the mission, not cutting anything. You can do it.
Come on.
Come on.
Come on.
Yes. I think we're clear.
Final hill climb.
>> [music] >> Can it do it?
Come on.
Wow.
First time up the hill. Overall, I'd say, given the test conditions, given what it I'm asking of it to do, did really well.
Up to now, most of my attention has been focused on capability. What I really haven't explored is how this machine is intended to operate. This small residential property isn't much of a challenge for the Navimow, but it may help answer some questions that the other test areas could not. All right, guys, we're down in the city.
It's noisy. It's crowded. We've got these really cool houses from probably the '50s or '60s, and we got a proper residential lawn for the Navimow.
I'm excited. I brought out a temporary setup, the docking station, the antenna, and we're going to map. We're going to do some mapping. We've got a side sidewalk, excuse me, in between areas.
So, yes, this is not like a 4-acre job, but I think it's fair to give the Navimow an opportunity to excel in the area that I think it was built to excel at. We'll connect everything to Wi-Fi, and then we'll start mapping. Like I said, we got this area here, a sidewalk, another area, and feeling good about it.
Do not cut overgrown grass or weeds. The exact thing I've been trying to do up in the mountains.
So, it's going. I want to see if I can adjust the deck height.
>> All right, so the Navimow.
It's meant for grasses like this.
Typical classic residential stuff. The mapping process was super simple. I set up this area, I set up the next area, and it just did what it was supposed to do. I was able to adjust the cutting height while it was cutting. Not all mowers you can do that with. So, that was pretty neat. And it just gave me a taste of how refined it is in this space.
I came into this property expecting to learn a little more about the mapping system. What I was not expecting was how different the experience would be once the machine had been given enough room to operate for a sustained amount of time.
At this point, I'm I'm still not sure where the review is going to end up, but I do have a feeling that there's more to this machine than I've seen so far. All right, we reached our starting point.
The general idea is I'm going to map this whole area out.
The grass has already been cut down to about 3 in or so.
We're going to set this at 2 in probably, and number one, see how it does on terrain.
Number two, uh test the mapping in a larger area.
And number three, hopefully, if we run it slow, the cutting power.
>> Nice.
Nice.
And this is why you take a machine, motorcycle, a e-bike, a robo lawnmower to several different places to test it because here it really impressed me, whereas before up by the mountain house, we're still in the mountains, but at my place it was like, "Ah, I don't know." Uh I mapped out the border and it just I don't know why it chose the path it did, but it was relentless. 2-in uh blade height and it cut things that I did not think it was capable of cutting and it did multiple passes over problem areas.
It got itself unstuck several times. The front wheels being able to maneuver inside outside uh zero turn it was almost decision-making, which was really cool.
And to watch it work and to watch it finalize and actually cut that area down really gave me uh a better impression of what the machine is capable of and that's why we're out here.
After the mountain property, I found myself looking at the machine in a different light.
I had a new hypothesis. The question becomes will the hypothesis hold up when I bring the machine back to a familiar environment, my place, and allow the machine to operate on its own? At the risk of making this video too long, I'm doing one more test. So, the arc of this sort of video follows a little bit of skepticism, a little bit of I'm not sure if it's capable, and then the last test where we let this run by itself shocked me. I was like watching it do work and actually cut well and I was just like, have I been wrong this entire time?
So, we're back here at the property where we started and I'm going to let it cut the top of the septic system.
Mostly reasonably okay grass. There's some obstacles in the way.
But, I really really want to give it one more opportunity to work by itself. I think that is where it shines. That is where it makes decisions. That is where it truly puts in the work.
>> I don't understand why it makes the decisions it does to choose to mow how it mows.
You would think it would just go up and back, up and back, but no, it's making small little circles and going forward, going back, and turning, and coming back to an area.
I find that fascinating and I want to know more.
When I started this review, I thought that I was going to go ahead and film and test a robot lawnmower. What I eventually came to realize is that I was evaluating an autonomous system. The earlier tests helped me understand the machine, but the later tests really helped me understand how the machine was [music] intended to operate.
And those two turned out to be very different things. Every conclusion I reached throughout this video was based on the information I had available at the time. Some of those conclusions held up, others changed as additional testing provided new information. Looking back, the biggest lesson here wasn't about slope capability, cutting power, obstacle avoidance, or mapping. It's understanding that many of the things that impressed me only became apparent when I stopped trying to control [music] the environment and allowed the machine to do the work it was intended to do.
And that's exactly why this review took the path that it did.
Guys, thank you for watching, taking the time to watch. This has been a long extended review, [music] but I love doing it. I love chasing all of the um outcomes at every test location in order to formulate a final [music] opinion. The Navi Mo has been an incredible machine to work with. Sounds like a coworker.
Has been an incredible machine to test, [music] evaluate, review.
I have loved making this video. Not all Mo review videos are going to be like this, obviously, but this machine had me chasing what I thought it [music] was, and I found it towards the end. So, please leave a comment. Let me know what you think.
And uh I look forward to making the next one. Thanks, guys.
Related Videos

Setting up a curved screen with Immersive Calibration Pro 4 and multiple cameras (P3D v4)
FlyerOneZero
23K views•2019-07-21

Robot Learning with Sparsity and Scarcity
allenai
379 views•2025-10-14

Jorge Mendez-Mendez: Unlocking Lifelong Robot Learning With Modularity (2023-10-05)
umassmlfl
237 views•2024-01-06

Northwestern’s MS in Robotics: Student Robotics Projects, 2023
NorthwesternEngineering
1K views•2024-05-31

"Perfect" Turns: Turning by the Gyro - FIRST LEGO League (FLL) SPIKE Prime + EV3 RePlay Programming
ZacharyTrautwein
94K views•2020-10-02

Gorkem Secer: TSLIP-based Deadbeat Running Control of Bipedal Robot ATRIAS
DynamicWalking-wv6qm
298 views•2018-06-22

Self-Driving Cars Need Lessons On Human Drivers | Maddie About Science
skunkbear
26K views•2018-08-21

Milrem Robotics’ THeMIS UGVs used in a live-fire manned-unmanned teaming exercise
MilremRobotics
99K views•2021-05-20
Trending

we're almost finished the house (ep.125)
JennaPhipps
347K views•2026-07-22

We Finally Know Where Saturn’s Rings Came From
astrumspace
79K views•2026-07-22

BIG BET: Cathie Wood goes ALL IN on Elon Musk
FoxBusiness
89K views•2026-07-22

MIC DROP: Smithsonian Director Called Out For Woke Propaganda
TheAmalaEkpunobi
37K views•2026-07-23