How to set up a map on a robot vacuum cleaner Xiaomi Mi

Modern models of cleaning robots from Xiaomi and the Mijia ecosystem have advanced navigation systems that allow you to build accurate digital copies of your home. The correct configuration of the map is the foundation for the effective operation of the device, because it is based on the algorithms plan the routes and calculate battery consumption. If you ignore this step, the robot vacuum cleaner can randomly move around the apartment, skip lots or get stuck in the same places.

The process of creating and editing a map in the Mi Home app seems simple, but it contains many nuances that affect the quality of cleaning. Users often face situations where the device can not correctly determine the boundaries of rooms or mistakenly combines a corridor with the living room. Understanding the principles of work LDS-Navigation or visual positioning systems will help you avoid typical errors and get the perfect result.

In this article, we will go through all the steps from initial launch to fine-tuning virtual walls and exclusion zones. You will learn how to fix scan errors, divide rooms for more accurate statistics, and why you can not move the base while you work. Proper calibration of the system will allow your gadget to work independently for months, requiring only occasional emptying of the dust container.

Primary scanning and preparation of premises

The first launch of a robot vacuum cleaner to build a map is a critical step, on which the accuracy of all subsequent cleaning depends. Unlike the chaotic movement characteristic of budget models, Xiaomi devices with a laser rangefinder or camera must create an accurate floor plan.

Launch the device through the Mi Home app and select mapping mode. The robot will circle the perimeter of the room, scanning the space. At this point, it is strictly forbidden to open or close doors, rearrange furniture or lift the device on your hands. Any interference will knock down the coordinate system, and the SLAM algorithm (simultaneous localization and map construction) will give an error or create an incorrect projection.

⚠️ Warning: If you force the cleaning to stop during the first scan, or the robot loses its battery, the card may not be saved or be defective, and then it will have to be removed and the process restarted in Quiet Cleaning mode or a special map-building mode.

Once the cycle is complete, the device will return to base, and the app will show a notification to save the new map. Now you can proceed to edit it. Make sure that the lighting in the room was sufficient, especially if you use a model with visual navigation, because in total darkness, the camera may not recognize the textures of the floor and walls.

Room division and renaming of premises

Once the base map is created, the system often perceives the entire space as one large room or mistakenly combines adjacent spaces such as the kitchen and living room. To properly work the cleaning scripts and virtual walls, you need to separate. In the Mi Home application interface, go to the map settings and select the "Separate" tool.

Draw a separation line across walls or doorways. Virtual partitions need to be clear and pass through bottlenecks so that the algorithm can clearly understand where one zone ends and another begins. Don't divide rooms too small, otherwise the robot will waste extra time circumventing non-existent boundaries.

Once separated, give each site the correct name from the list (Kitchen, Bedroom, Corridor) and it's not just a cosmetic function: renaming allows you to give voice commands to a smart speaker, such as "Alice, ask the robot to clean the kitchen," and cleaning statistics will be kept separately for each zone, which will help you understand where pollution is happening more often.

If the automatic separation failed, use the “Combine” tool to erase the extra lines.

Configuring virtual walls and prohibition zones

One of the most useful features of Xiaomi’s ecosystem is the ability to create invisible barriers: Virtual walls and exclusion zones allow you to fence sensitive areas of the floor, places with wires or areas where there is a bowl of water for pets. There are two main types of restrictions: an invisible wall (a line that cannot be crossed) and a exclusion zone (a square or rectangle where entry is forbidden).

To adjust, go to map editing mode and select the appropriate tool. You can draw a line along the threshold or highlight a square around the legs of the chair. The exclusion zones are especially relevant for models without brush lifting, since they prevent carpet wet or fringe entanglement.

Settings → Card Management → Virtual Restrictions

Use this tool wisely: too many virtual walls can confuse the navigation system, forcing the robot to choose suboptimal routes. Try to make the exclusion zones as minimal as possible. It is also worth noting that virtual walls are tied to the orientation of the base, so when it moves, the settings can get lost.

💡

Use a no-go zone of at least 0.5x0.5 meters, and the robot may ignore too small areas due to positioning error or movement inertia.

If you use Quiet Cleaning at night, set up a separate restriction profile that prevents you from entering bedrooms so that noise doesn’t interfere with sleep, allowing you to flexibly manage your device’s behavior at different times of the day without physically moving furniture.

Keeping several maps and working with floors

Multi-card storage will be useful for owners of multi-storey buildings or apartments with complex layouts. Xiaomi flagship models (for example, the S10, S12, X10 series and later) support storing up to 4-5 different cards in memory, which means that you can remove the first floor, move the base to the second, build a new map there and save it.

When you move the base to another floor, the robot must rescan the space. The app will suggest saving the current map as new or updating the existing one. Choose Save as new if the floors are different layout. When you start cleaning on a new floor, the device automatically recognizes the location by characteristic features and activates the corresponding floor map.

It's important to understand that models with gyroscopic navigation (usually budget segment) are not able to build detailed maps and save them. They work on the principle of "snake" or chaotic, so the multi-storey function is not available for them. For such devices, the only solution is to buy an additional base for each floor, although this does not guarantee perfect work without a map.

💡

Multi-mapping is only available in models with laser (LDS) or advanced visual navigation. Budget gyroscopic models do not save the map.

Problem Solving: Map is lost or not built

Often users are faced with a situation where the robot begins to “get lost”: the map in the application twitches, double walls appear, or the device spins endlessly in place. The main reason for such failures is contamination of the sensors. The laser rangefinder (turret on top) and the lower cliffs sensors (cliffs) require regular cleaning with dry soft cloth.

Another common problem is poor lighting, or direct sunlight hitting sensors. If the robot is operating in total darkness, the visual models may be wrong. Also check if the wheels are stuck in the carpet pile or wires. If the robot was raised while it was working, it needs to be returned to base to restart the coordinate system.

⚠️ Warning: Never attempt to manually deploy the robot or move it to another room while building a map.This is guaranteed to desynchronize data and require a complete map reset.

If errors repeat, try resetting the map completely through the app menu and starting the construction again, before clearing the floor of foreign objects as much as possible. Sometimes it helps to update the firmware of the device, in which developers improve the algorithms for processing data from the sensors.

Sensor maintenance for accurate navigation

The accuracy of the map depends on the purity of the optical and laser elements. Once a week, wipe the laser rangefinder protective glass and sensors on the bottom. Dust, animal hair and fat plaque (relevant to the kitchen) can distort the readings, making the robot think that a new object has appeared in the room.

Pay special attention to the side brush and the main roller. If they are wound hair, the balance is broken, and the robot can move to the side, which also affects the construction of the map. Regular maintenance is the guarantee that the digital twin of your apartment will always coincide with reality.

Don’t forget to check the software version in the Mi Home app. Developers regularly release patches that improve obstacle-avoidance algorithms and positioning accuracy. Ignoring updates can lead to less efficient cleaning scenarios.

Frequently Asked Questions (FAQ)