Modern cleaning devices like the Xiaomi Mi Robot Vacuum have impressive intelligence, but they require precise navigation to work efficiently. Map building is a fundamental process that turns chaotic movement of the machine into a systematic and efficient cleaning. Without a properly designed floor plan, the robot will not be able to use advanced features such as dividing into rooms or creating virtual no-go zones.
The process of creating a digital twin of your home takes time and requires proper preparation of the room. Many users make the mistake of running the device at full power in a cluttered space, leading to navigation errors and "stuck." A laser rangefinder mounted on the housing must freely scan the environment so that the algorithms build the correct geometry of space.
In this article, we will discuss all the steps in detail: from preparing an apartment to fine-tuning the saved plan in the Mi Home application. You will learn how to avoid common mistakes during the first launch and how to make your gadget clean exactly where it is necessary, ignoring dangerous areas for technology such as long-pile carpets or places with scattered wires.
Preparation of the premises before the first launch
Before Xiaomi's robot vacuum cleaner can build a map, it needs to be given the perfect scanning conditions. Remove any small objects that could be pulled into a turbobrush or hit by the device's body from the floor. Wires, socks, toys and dangling curtains are the main enemies of accurate positioning and successful mission completion.
Special attention should be paid to lighting and floor conditions. Although many models are equipped with LiDAR navigation that works in the dark, the lack of light can make it difficult to work additional optical sensors to prevent falling from height. In addition, wet floors or overly shiny tiles can distort the readings of the sensors, so it is better to wait for the surfaces to dry completely after washing.
⚠️ Warning: Do not attempt to build a map if the room is running active sources of IR radiation, such as open fireplaces or powerful heaters pointing to the floor.This can blind sensors and lead to a distorted plan with holes".
It is also important to check the base: the charging station should stand on a flat hard floor, preferably against a wall, with free space 0.5 meters on the sides and 1.5 meters in the front. If the base is standing on a carpet or in a niche, the robot may not find its way home after completing the first test detour.
The process of primary scanning and mapping
Once the room is ready, you can go to the scanning process itself, run the device through the Mi Home app or button on the case, selecting the normal cleaning mode, at which point the robot vacuum cleaner will start a chaotic movement, exploring the perimeter of the room and identifying obstacles, do not interrupt this process, even if you think the device is driving pointlessly.
The first run can take much longer than a normal cleaning, as the algorithm builds a virtual grid of coordinates. The device will repeatedly return to the areas already studied to clarify the data received from the laser rangefinder.
If you have a large apartment, you may not have enough charge to scan it completely. Modern Xiaomi models can go back to the base to recharge and continue building from the stop, but for the first time, it is better to ensure maximum battery power by opening all interior doors so that the device can cover all available area at once.
☑️ Ready for scanning
Configuring and editing a saved plan
When the initial scan is complete, a saved map will appear in the Mi Home app, but it may not look perfect: walls may be eaten and space is not divided correctly, and you need to enter the editing mode that is usually available after cleaning or in the map settings to correct the situation.
The first step is often to split or merge rooms, and if the robot mistakenly combined the kitchen and living room into one huge room, use the "Separate" tool, draw a line along the actual wall, and the algorithm will divide the area into two independent parts, which will allow them to set different cleaning modes in the future.
The opposite is true when the corridor is broken up into many small pieces, and the Unite tool will help you glue them together into a single logical zone, which is especially important for the proper operation of the Room Cleaning function, where you want to send the machine to a specific room only, not to the entire apartment at once.
⚠️ Attention: When manually editing walls, try not to make lines too close to real obstacles. 5-10 See that the robot does not try to "wipe" its border close to the legs of furniture and not damaged them.
Creation of virtual walls and forbidden zones
One of the most useful features after mapping is the ability to create virtual constraints, and unlike the physical magnetic tapes that were used in the old models, here you draw zones right on your smartphone screen, which is indispensable for protecting places where wires are scattered, where there are bowls of water for animals, or where fragile objects are located.
There are two main types of restrictions: a "Virtual Wall" (a line that cannot be crossed) and a "No-go Zone" (a square or rectangle that cannot be entered). For long-pile carpets that are not recommended for washing models, a square no-go zone is ideal. A linear virtual wall is better to restrict access to another room or to the children's arena.
So you need to be careful about setting up these areas, and make sure that the boundaries that you draw are exactly the same as the actual situation on the floor, and if you keep pushing yourself against the barrier that you're not seeing, you can go into error mode, or you can just turn around and leave large areas of usable space undisturbed.
What if a robot ignores virtual walls?
Keeping several maps for different floors
Owners of multi-storey houses or two-level apartments are often wondering if it is possible to save a second-floor map. Xiaomi Mi Robot supports storing multiple cards (usually up to 5 pieces), but switching between them requires the right sequence of actions. You can't just move the device to another floor and expect instant recognition.
To create a new map, you first need to clear the memory of the current one or create a new saved copy. If you move the robot to the second floor, start building the map again, as you did when you first turned on, the device will create a new layout file that will be saved in the list of maps of the device.
If you clean the app regularly, you’ll have to manually select the map you want in the app before you start, or use automation scenarios if your model and software version support it, and ignoring this step will cause the robot to try to match the current environment with the ground floor map, causing chaotic behavior and navigational errors.
| Parameter | Description | Recommendation |
|---|---|---|
| Scanning area | Maximum area of one card | Up to 200-250 m2 without tears |
| Number of cards | How many plans are stored in memory | Usually up to 5. |
| LiDAR accuracy | Distance measurement error | ± 2-3 cm under ideal conditions |
| Time for construction | Duration of the first run | Depends on the area, 1.5-2 times longer than usual cleaning |
💡
To make it easier to switch between floors, create scripts in the Mi Home app: "If I'm home on the second floor -> Download the 2nd floor map." This will save time for manual adjustment.
Frequently encountered problems and their solution
Even if you follow all the instructions, you can have a lot of problems, and one of the problems that you often have is you drive around, and the plan of the room gets distorted or shifted over time, and this is often because the robot has been raised and moved while you're working, or the base has been shifted, and in these cases, it helps to completely reset the map and build again.
Another problem is that you can't find a base after cleaning, and if the map is built right, but the robot is spinning around the station and it doesn't dock, check if the charging contact is blocked, and clean the sensors on the robot and on the base, and also make sure that the base is not in the sun, because the infrared sensors can go blind from the direct rays.
Sometimes the app shows that the card is saved, but when you start cleaning, the robot behaves as if there is no card. In this case, it helps to reboot the device itself (turn off by the switchboard for 10 seconds) and reconnect to the Wi-Fi network. It is critical to use a 2.4 GHz Wi-Fi network, since most models of Xiaomi robot vacuum cleaners do not support 5 GHz frequency, which can cause problems with communication and card transfer.
⚠️ Warning: Never try to wash a laser rangefinder (tower at the top) with a wet cloth while working. Getting moisture inside the rotational mechanism will result in expensive repairs. Use only dry microfiber.
Optimizing cleaning with zones and routes
Once the map is built and edited, it's time to fine-tune it, so you can set up the order of the rooms so that the robot can clean the faraway spaces first and then move to the exit, and this prevents the floors from being recontaminated as they move to the base.
Use Area Cleaning for local cleaning, and if you drop something on the floor at a particular point, just draw a rectangle around that spot on a map, and the robot will arrive there, ignoring the rest of the apartment, saving battery life and time.
Regularly update the device software. Xiaomi engineers are constantly improving the algorithms for mapping and navigation. New firmware can significantly improve the positioning accuracy of your device and add new mapping features that were not available before.
💡
A well-built and edited map is 90% of the success of a self-driving cleaning.Spend time on initial setup to enjoy cleanliness for years without human intervention.