Real-time positioning method of mobile robot and mobile robot

Abstract

The invention discloses a real-time positioning method of a mobile robot. The method includes the steps of fusing speedometer information and gyroscope information into positioning information of the mobile robot, conducting the SLAM algorithm through the combination with scene information obtained through a binocular camera with the positioning information as the predicted value of the SLAM algorithm, and achieving positioning. By means of the method, the positioning information of the mobile robot can be corrected in real time, accumulated errors of a speedometer can be effectively reduced, the positioning accuracy is improved, and accumulated errors of a system are reduced. The invention further discloses the mobile robot using the method, and particularly relates to a sweeper for cleaning floors.

Description

Technical field [0001] The invention relates to a real-time positioning method of a mobile robot and a mobile robot. Background technique [0002] With the improvement of human intelligence, countries all over the world attach great importance to the development and research of autonomous mobile robots. For example, indoor sweepers are recognized by more and more people due to their intelligence and convenience. Complexity and variability, this requires it to have real-time perception of the environment, judgment of obstacles and other related capabilities, to achieve its own positioning and map creation. In addition, due to the fact that the wheels of the indoor sweeping machine in the middle shift will slip and slip, the positioning accuracy of the odometer will have a large cumulative error over time. [0003] At present, indoor sweepers are mainly divided into two categories: [0004] The first is a traditional sweeper with random cleaning, which mainly uses relatively inexpens...

AI Technical Summary

Problems solved by technology

Among them, inertial navigation is based on Newton's laws of mechanics. Robots can rely on inertial navigation to realize autonomous three-dimensional positioning, calculate pose and measure speed without any optical, electrical, magnetic and other information contact with the outside world. Before the first use, a long initial alignment time is required, and the navigation information is generated through integration, the positioning error increases with time, and the long-term accumulation results in poor accuracy; the laser radar measurement accuracy is high, and the measurement data can be used directly to assess the environment. Feature extraction, but due to its high cost and complex structural design, its scope of use is also limited; visual sensors can provide rich scene information, wide detection range, high spatial resolution, and monocular vision Sensors have been widely used in industrial robots, mobile robots and other fields, but the existing monocular vision robots cannot directly obtain the depth information of the environment, and need to add additional algorithms to realize the ability of environment recognition. The calculation is more complicated and the accuracy is lower. Can not better meet the real-time positioning

[0006] The above two sweeping machines have not solved the problem that the positioning accuracy of the odometer of the mobile robot, especially the sweeping machine, will have a large cumulative error over time.

Images