Binocular vision positioning and three-dimensional mapping method for indoor mobile robot

Abstract

The invention discloses a binocular vision positioning and three-dimensional mapping method for an indoor mobile robot. A binocular vision positioning and three-dimensional mapping system for the indoor mobile robot is mainly composed of a two-wheeled mobile robot platform, an odometer module, an analog camera, an FPGA core module, an image acquisition module, a wireless communication module, a storage module, a USB module and a remote data processing host computer. The FPGA core module controls the image acquisition module to collect information of a left image and a right image and sends the information of the left image and the right image to the remote data processing host computer. Distance information between images and the robot is obtained on the remote data processing host computer according to the information of the left image and the right image in a fast belief propagation algorithm, a three-dimensional environmental map is established in the remote data processing host computer, recognition of a specific marker is realized in an improved SIFT algorithm, and then the position of the mobile robot is determined through information integration in a particle filter algorithm. The binocular vision positioning and three-dimensional mapping system is compact in structure and capable of having access to an intelligent space system so that the indoor mobile robot can detect the environment in real time and provide more and accurate services according to the corresponding position information at the same time.

Description

technical field [0001] The invention belongs to the technical field of indoor mobile robots, and relates to a method for binocular vision positioning and three-dimensional mapping of an indoor mobile robot. Background technique [0002] The first focus of indoor mobile robot research is environmental perception, and sensors are an important means for robots to obtain environmental information. There are many types of sensors carried by mobile robots, and those related to positioning can be divided into two categories: internal sensors and external sensors. Internal sensors refer to sensors used to perceive the robot's own position and motion parameters, such as gyroscopes, photoelectric code discs, Doppler meters, etc. These sensors can be used to estimate the robot's own position. Among them, the robot odometer based on photoelectric code disc and dead reckoning method is the most commonly used internal sensor for mobile robots because of its low price, high sampling rate,...

AI Technical Summary

Problems solved by technology

Use processors with stronger data processing capabilities to execute complex algorithms to further solve problems with a large amount of calculation and improve real-time performance

Images