With the advancement of science and technology and the improvement of people's living standards, many families now own an intelligent robot vacuum. However, not all robot vacuums are equal. Among them, an important technical ability that affects the cleaning effect of robot vacuums is their navigation system.
Navigation ability can be said to be the essential basis for the robot vacuum to complete cleaning tasks. It is strongly related to how the robot vacuum recognizes the surrounding environment, plans its cleaning route, ability to avoid obstacles, and cleaning efficacy. The navigation technology of robot vacuums has evolved from the unplanned and exploratory navigation in the early stage to the level of actively and meticulously planned navigation. So, presently what are the main types of navigation technology of robot vacuum?
The navigation system of the robot vacuum is generally divided into three types, inertial navigation, laser navigation, and visual navigation.
- Inertial navigation
Inertial navigation means that the robot vacuum uses built-in gyroscopes, accelerators and other sensor devices to measure the angular acceleration and linear acceleration, and then obtains the positional information of the robot through calculations. The accuracy of the measurement is subject to the gyroscope drift, calibration errors, sensitivity and other factors, so the accuracy is lower, and the error will continue to increase with the increase of use time. It is more suitable for small areas and simple environments. It is categorized as a relatively low level of active planning navigation.
- Visual navigation
Visual navigation is easier to understand, that is, by equipping the robot vacuum with a camera to simulate human vision to realize the recognition and navigation of the surrounding environment. Visual navigation mainly uses two kinds of vision sensors to obtain information. One is a depth camera, which realizes three-dimensional space perception through distance measurement. It is also regarded as a distance measuring sensor, which is an active light source distance measuring sensor, including structured light and phase TOF. The second is the binocular, multi-eye, fish-eye navigation sensor, which is a non-active light source sensor. The working mechanism is similar to that of the human eye. According to the principle of triangulation ranging, the distance information is calculated by analyzing the difference between the images collected by the two sensors.
The disadvantages of visual navigation technology are clear. The hardware has many operating weaknesses. It relies on its camera to collect information. The efficiency of the camera and distance sensor is susceptible to interference from ambient light. In the case of poor lighting conditions, it can hardly work. Visual navigation technology requires very good lighting for it function accurately. Secondly, the non-active light source distance/range sensor functions like the human eye. The greater the distance, the greater the error. At the same time, too much light also affects the processing unit causing it unable to calculate distance and range accurately.
- Laser navigation
The basic principle of laser navigation technology is relatively simple. Laser ranging is to emit a beam of light toward a specific direction, and the light bounces back when encountering an object to be captured by the receiver. The speed of light is known, and the transit time is known. Then you can calculate the distance between yourself and the object. The difference is that laser ranging only needs to be launched once and received once.
While laser navigation technology is one dimension higher. By measuring more points in various directions, you can build two-dimensional maps or three-dimensional modeling, and at the same time determine the robot vacuum's own Location;
The second is to use the distance information between two points to perform triangulation distance measurement, obtain more accurate distance information according to the triangulation distance measurement or TOF algorithm, and finally generate a map model of the surrounding environment. Accordingly, robot vacuum further plans the cleaning route. The measurement accuracy of laser navigation is obviously much higher than that of inertial navigation, and the map resolution is also very much higher. It is currently a popular navigation solution used by most robot vacuums in the market.
Many products in the market now use the vSLAM laser navigation algorithm. For example, the flagship model of the Uoni robot vacuum V980 adopts the laser navigation method, which provides more accurate mapping and smarter obstacle avoidance functions. With the vSLAM algorithm system, it can quickly scan, memorize and plan the home cleaning path of the robot vacuum without getting stuck, sleepy, or missing spots.Truly achieving fast, accurate, and thorough cleaning effects.