A Research On In-door Service Robot Navigation Technology Based On Binocular Vision

With the development of computer technology and robotics related technologies, research and development of service robots has been widespread concern. The navigation technology was one of the key technologies, and was also the critical intelligence and autonomy to achieve service robot technology. It could be said that the development of service robots is largely driven by the development of navigation technology. Therefore, the research and development of service robot navigation technology could get great help for future service robots.The development status of service robot navigation technology in recent years has been studied in this paper, as well as the future trends. Comparing each navigation system, the visual navigation with a high level of their intelligence as well as information and other features was a study trend of future service robot. Having the similar structure of the human eye, binocular vision navigation system was a research trend in vision navigation. Therefore, the direction of this thesis indoor service robot navigation based on binocular vision was determined in this paper. The two key technologies in binocular vision navigation- binocular vision targeting positioning and path planning were expanded. The basic principles of camera calibration principle and Zhang Zhengyou calibration method and procedure were first introduced in the binocular vision targeting positioning algorithm. Then the target recognition algorithms were briefly described, using geometric moment invariants combined with Euclidean distance goal recognition algorithm for image matching the shape of each region. Later the matching algorithm based on the weighted average algorithm with arctangent normalized of the three types of regional characteristics was proposed according to the paper requirements with the left image of the target area having been normalized after characterizing subtracting absolute value weighted by regional matching results. Finally, the recovery based on least squares method for three-dimensional information was researched. The existing path planning algorithms have been studied, and the path planning algorithm based on artificial potential field was analyzed deeply. For less than the traditional artificial potential field in the path planning algorithm were solved by adding regulators Destination Unreachable problem, using traversal algorithm to solve the problem of local minima introduced to overcome the obstacle azimuth traditional methods with reality phenomenon.In order to verify the correctness of the method proposed in this paper, in high-precision mobile platforms in binocular vision positioning accuracy verification test, and the movement of the experiment on binocular vision AGV mobile platform. Tests show that the improved algorithm herein, binocular vision positioning accuracy can reach 1mm, attitude measurement accuracy of 0.9 °. AGV platform test showed that binocular vision positioning system designed in this paper can be reliably work in the course of the campaign,.