Each robot vacuum we consider suggesting will be rhythmic in our test lab in Louisville, Kentucky. In addition to the test floor running controlled pickup tests, we monitor each robot vacuum in a special test room with simulated furniture to measure how much it navigates around common obstacles. In the past, we checked the capabilities of each robot vacuum cleaner without clogging or leaving loose bondage, we considered the wipe function, and we checked how it also navigates the fake dog mess.
Starting with our performance testing, let’s dig into the main considerations.
Robot vacuum rating index
Not clear
| Sub-rating category | weight | What are we looking for | |||||
|---|---|---|---|---|---|---|---|
| Performance | 30% | Performance scores inferred from AVG (AVG_SAND + AVG_BLACKRICE) | |||||
| Value/Price | 25% | Consider all other features, retail price ratings. That is, does this price seem fair to the value provided? Is this proven to be justified by performance, functionality and NAV efficiency? | |||||
| feature | 15% | What functions does vacuum provide? Self-empty station? Multiple batteries? NAV technology? mop? | |||||
| Runtime | 20% | Consider the navigation efficiency score (1-10) based on the time it takes to complete a full cleaning cycle in the CNET Labs custom robot vacuum NAV test room. | |||||
| Ease of use | 10% | UX-How easy is setting up the experience/Quikck? Does it come with smart home features? Smartphone app? Voice assistance? |
Robot vacuum pickup power
On the vacuum cleaner, we want to know how effective each robot is for common debris and other debris, and how it can fare with smaller particles like dust, dirt and sand. To find out, we used dried, uncooked black rice as a stand-in for bread crumbs and sand as a fine grain analog.
In each case, we spread a controlled quantity across three test floors: low purple carpet, medium wood carpet and hardwood floors. Lower carpets are shorter, plush carpets are shorter and fibers are shorter, so robot vacuums are easier to pick from (though not always). The mid-abdominal is softer, taller carpet with high fiber. For robot vacuums, it tends to be more challenging (though not always). We then vacuum the robot completely emptied the dust box, sent it to clean the affected area, and finally measured the weight of anything we managed to pick up. This gives us a full pickup percentage. From there, we repeat each run twice and average the results.
In recent months, we have canceled tests of black rice on hardwood floors because every robot vacuum cleaner we tested is 100% close to 100%. Now we use sand testing as a primary benchmark for evaluating cleaning performance. We think anything 50% or more is a good result of sand.
Hardwood Floor Test
Low-rise carpet test
Middle carpet test
These overhead long shots show traces of the Roborock S7 Maxv Ultra every time, as it cleans our test chamber. We attach the glow rod to the top of the detergent directly above the vacuum absorption to understand the coverage the vacuum provides and its smart navigation. In this case, the S7 MaxV Ultra is as thorough and consistent as a robot vacuum.
Robot vacuum navigation skills
Your robot vacuum can only thoroughly clean your home like navigation. The ideal cleaner will easily find your way from room to room and automatically avoid obstacles in the process, all with proper, low maintenance automatic cleaning.
We made sure to observe each robot vacuum cleaner as it can be cleaned to understand the level of its navigation, but for the best comparison from cleaner to cleaner, we made the best comparison of each robot’s highly exposed lens as it cleans the dark test chamber with a glow stick and attaches to the top of everyone above each of the vacuum absorbs. The light trajectory shown to us shows the path of the robot as it navigates the room and cleans our simulated furniture.
Now compare it to the next GIF which shows you three runs from the Irobot Roomba Combo J7 Plus. Pay attention to the difference? Roomba is less effective in covering the entire room, lacking the lower left corner on two-thirds of the run, and having enough coverage on the legs of that simulated dining table.
To a large extent, it depends on the role of technology. For years, we have pointed out that robot vacuums that use laser-guided lidar navigation are often very good at mapping the environment and finding their own way out. Meanwhile, a 3D mapping camera with object recognition intelligence can enable the robot to vacuum and adapt to obstacles in its path. Both of these technologies are used by the Roborock S8 Pro Ultra, which helps explain why it performs so well here. Meanwhile, Roomba relies solely on camera and sensors, while the laser is not mixed.
IROBOT ROOMBA J7 PLUS delivers on the promise to avoid shit (fake or otherwise).
Still, those cameras will certainly come in handy. Just watch the GIF above, which shows what happened when we tested the Irobot Roomba J7 Plus – especially its promise to identify and avoid pet waste. Roomba, with various (I assure you, fake) dog poops spread across a small, enclosed test floor, trying to vacuum the area without touching either. It succeeded and never touched any of our disgusting test poop.
Samsung Jetbot AI and our solid pet waste tests are always failing. In every run it ends up running over or pushing one of our shit models.
Now, it is expected to use its camera to discover and avoid pet poop compared to Samsung’s Jetbot AI Plus. The result is not very good. In each test, it ends up hitting one of our test heaps. Thankfully, they are not real.
