Xiaomi's modern robot vacuum cleaners are no longer just randomly moving discs that push against the legs of chairs. LDS-navigation (laser rangefinder) allowed these devices not just to clean, but literally "see" the room, creating an exact digital copy. It is this process, known as mapping, that turns routine cleaning into an intelligent process where each area is cleaned methodically and efficiently.
For many users, the initial setup seems complicated, but in fact, the Mi Robot Vacuum algorithm is quite simple and logical: you don't have to draw a plan of the apartment by hand or drive the device behind you like a radio-controlled machine. All you need is to properly prepare the room and start one full cleaning cycle, during which the laser sensor collects the necessary data on the geometry of walls and obstacles.
Once you complete the first pass, your smartphone will have a precise layout of your home that can be edited, divided into rooms and equipped with virtual boundaries, allowing you to access zoned cleaning and route building for multiple floors. Understanding how the navigation system works will help you avoid typical mistakes and make the equipment work at maximum efficiency.
Principles of operation of laser navigation and scanning
The basis of the exact positioning of Xiaomi robots is LDS-This is a sensor that is located in a characteristic elevation on the top of the device, and this module is constantly rotating, emitting laser beams that are reflected off walls, furniture and other objects. By measuring the time and angle of return of the reflected signal, the processor calculates the distance to the obstacles with millimeter precision, forming a point cloud in real time.
The data is processed by SLAM (Simultaneous Localization and Mapping) algorithms, which allows the device to simultaneously build a map and determine its location on it. Unlike devices with camera navigation or gyroscopes alone, laser-equipped models create a persistent map that doesn't get lost easily. This means that even if you interrupt cleaning and turn on the device in a few hours, the robot vacuum cleaner will know where it left off and continue working from the right point.
It's important to understand that the quality of the map depends on the scanning conditions. The laser beam should be free to pass around the perimeter of the room. If there are too many mirrors, black glossy surfaces or high-fat carpets in the room, the distance sensor can get distorted data, in which case the system may mistaken the reflection in the mirror for an extension of the room or fail to notice the black furniture, mistaking it for emptiness.
⚠️ Warning: Do not seal or block the rotating dome LDS-Any obstacle, even a thin thread from the curtains, that gets into the rotational mechanism can cause the scanner motor to break and cause an error "LDS blocked".
The condition of the wheels and sensors is also critical to the stability of the algorithms. If the robot is stalling on a doorstep or carpet, its perception of the path traveled is distorted. So before you first start to create a map, make sure that the wheels are clean and the sensors on the bottom (especially the elevation sensor and optical tracker) are dust-free.
Preparation of the premises before the first launch
The quality of the final map depends 80% on how you prepare the apartment for the first start. The ideal situation is an empty room, but in reality it is impossible. Your task is to minimize the number of temporary obstacles. Remove from the floor wires, toys, socks and other small objects that may become entangled in brushes or be mistaken for a stationary object.
Pay special attention to doors. To properly construct the geometry of rooms, all interior doors must be open. The robot must be able to move freely from one zone to another to understand their relationship. If the doors are closed, the device will build several isolated maps or one large map with "deaf" zones, which then have to be manually combined or edited.
You need to set up a charging station in advance, it should be against a wall, with 0.5 meters of free space on the sides and 1.5 meters in front, don't hide the base in a niche or behind curtains, the laser needs a view. Once the base is installed, let the robot fully charge before the scanning procedure begins.
☑️ Preparation for mapping
It's also worth checking the light level. Although the laser rangefinder works independently of light, optical sensors (used in some models for precise positioning) may work worse in total darkness or in bright, direct sun, creating glare.
The process of primary scanning and mapping
When the room is ready, run the full cleaning cycle through the Mi Home app. Don't use Point Cleaning or remote control for the first card. The robot must travel the entire perimeter so that the algorithm can close the contours of the walls. As it moves, the device will move around the perimeter and then fill the center of the room with zigzag movements.
At this time, you'll see a real-time mapping in the app, gray lines showing the walls that have already been scanned, and a blue fan representing the current scanning sector. LDS-If the robot encounters an obstacle, it gently walks around it and continues to build the map further. 50-60 square meters takes up about 40-60 minute.
If the robot is stuck, the app will notify you, and when released, it will try to locate and continue, and interrupting the process in the middle (e.g., battery drain) will cause the card to be partially retained and the procedure will have to be repeated.
What to do if the robot gets lost?
Once the cycle is complete, the card will automatically be stored in the device’s memory and the cloud, and you can now move on to fine-tuning it, renaming rooms, and setting limits.
Editing the map: division and unification of rooms
Automatic room recognition works well, but it's not perfect. Often, an algorithm can combine the kitchen and the living room into one space, or, conversely, divide one large hall into two because of a high threshold or carpet. To fix this, there's a map editing mode in the Mi Home app (or Xiaomi Home).
To divide the room, select the Split tool, and you'll need to draw a line through the area you want to separate, and the robot will divide the room along that line, and you can assign each part a separate name, like "Kitchen" and "Distanting Room," which will allow you to set different suction power modes for different types of coatings in the future.
If the opposite happens, and one room breaks into fragments, use the Merge tool, and when you single out adjacent areas, you put them together into a single circuit, and after all the manipulations, remember to press the Save button, otherwise all the changes will be lost the next time you run.
- 🏠 Rename: Give rooms clear names ("Bedroom", "Corridor") to make voice control work properly.
- 🔄 If the orientation of the map does not match reality, use the 90 degree rotation function.
- 🗑️ Removal: Unused areas (such as a balcony where the robot should not go) can be removed from the map completely.
- 💾 Save: Any changes will only take effect after confirmation and saving to the cloud.
Proper zoning is key to flexible cleaning, so you can only send the robot into the kitchen after lunch or into the hallway after walking your dog without having to clean the whole apartment.
Configure virtual walls, zones and forbidden areas
One of the most powerful features of mapping is the ability to create virtual constraints, and unlike physical magnetic tapes that need to be glued onto the floor, virtual walls are drawn directly on a map in an app, which allows you to instantly deny the robot access to certain places.
There are several types of constraints. Virtual walls (red line) create an insurmountable border, so that the robot will not cross it under any circumstances. This is ideal for enclosing areas with wires or places where there is a bowl of water for animals. No-go zones (red square) indicate areas where you can not even go partially, such as around the legs of chairs or complex equipment.
You can also have invisible walls for specific modes, and you can set the map so that when you dry cleaning, the robot will come on the carpet, and when you wash the floors, it will go around it so that you don't get the pile wet, and this feature is called "Zones without a mop mat," or similarly, depending on the model.
| Type of restriction | Visualization | Function | Example of use |
|---|---|---|---|
| Virtual wall | Red line | Complete intersection ban | Border between rooms, entrance to the balcony |
| The no-go zone | Red square. | Prohibition of entry inside the perimeter | Around pet bowls, complex designs |
| Carpet-free zone | Blue square/line | Ignoring when washing | Carpets with long pile, electrical engineering |
| Cleaning area | Green square. | Priority cleaning | A place under the table where they often crumble |
Using these tools allows you to adapt the behavior of even the smartest robot vacuum cleaner to the specifics of your apartment, preventing getting stuck and spoiling things.
Multi-storey maps and keeping multiple plans
Owners of homes or two-level apartments will benefit from the multi-card saving feature. Xiaomi's flagship models support up to 4-5 different cards in memory. This means that you don't have to reconfigure zones and walls every time you move the robot to another floor.
To activate this feature in the map settings, select Save Map or Add New Floor. Once the robot has built a ground floor plan, save it. Then move the device to the second floor, run a new scan, and save the result as Floor 2. In the future, the robot will automatically recognize which floor it is on by characteristic features (base location, walls), or you can choose the floor manually in the application.
⚠️ WARNING: When moving the robot to another floor to build a new map, be sure to turn off the Auto-Recover Map feature or clear the current map if you want to create a completely new plan rather than complement the old one. Otherwise, the robot will try to pull the new geometry onto the old map, creating chaos.
Keep in mind that switching between cards takes time to recalibrate the sensors. Don’t expect to immediately start cleaning immediately after switching on the new floor – give the device 1-2 minutes to navigate.
💡
Multi-mapping is not available on all models, and budget versions with gyroscope navigation often fail to save more than one room plan.
Typical problems in construction and their solution
Despite the advanced technology, users can experience errors, and one of the common problems is shifting the map, which happens when the robot is forcibly moved during cleaning or if it has been towed for a long time, and as a result, the robot can be in the wall or outside the apartment on the screen of the smartphone.
Another problem is split rooms or phantom walls, often caused by dark objects (black chair legs) that the laser can't see, or mirrors that create a false extension of the room, in which case manual editing or using the Restore Map feature, if available in your firmware version, helps.
If the robot is permanently losing contact with the base after building a map, check if the furniture arrangement has changed. The navigation system is tied to stationary objects. If you move the sofa or remove a large chair, the robot may become disoriented. In this case, it is recommended to update the map by starting the full cleaning cycle again.
- 📉 Weak Wi-Fi signal: If the robot loses communication in the middle of the room, there may be a bad signal.
- 🪞 Mirrors and Glasses: Close the bottom of large mirrors or use virtual walls to enclose glass doors.
- 🌑 Dark furniture: Black objects the robot may not see. Protect them with no-go zones.
- 🧶 High thresholds: If a robot gets stuck, use virtual lines to prevent it from passing through a specific threshold.
Regular maintenance of sensors and software updates through the Mi Home app helps to minimize errors and keep navigation algorithms up to date.
💡
If the map often “flys” or shifts, try to clear the history of the cards in the application settings and recreate them from scratch in conditions where there are fewer people and animals in the apartment.