Path control of panoramic visual recognition for intelligent robots based-edge computing

Abstract

The application of service-oriented intelligent robots in hospitals solves the problems of manual service such as delivery of medicines and orders from the medical aids. However, there is a problem of robot path planning, which leads to poor service. Taking the self-control robot with visual recognition capability in medical institutions as the research background, this paper proposes an image edge detection algorithm based on three-dimensional features. First, reconstruct the three-dimensional image through visual recognition, and calculate the three-dimensional feature map of the scene in the hospital. Secondly, the edge computing algorithm is used to realize the route area diversion, and the best route area is planned in the 3D feature map to realize the delivery of medicines and other items in the complex environment of the hospital, so as to achieve the purpose of improving the technical level of medical auxiliary equipment. Finally, the image restoration technology can improve the image clarity of the robot vision system and the robot automation level, and it is also helpful for the promotion of robot automatic path planning and control technology. The experimental results show that, compared with the hidden Markov algorithm, the method in this paper is significantly better than the hidden Markov algorithm, which meets the needs of medical institutions for intelligent robots.