Design of Blind Guiding Robot Based on Speed Adaptation and Visual Recognition

With the continuous increase in number of visually impaired individuals, it has become urgent to address traffic challenges faced by this group. A blind guiding robot based on speed adaptation and visual recognition was designed problem. The person is achieved through feedback control distance speed. Traffic signals are identified using optimized method YOLOv5 transfer learning, man-machine interaction realized applying multi-module units such as real-time image, speech, positioning. experimental results show that deviation between relative man machine set controlled within 13.1%, velocity 0.3 m/s, accuracy identifying reached 91.88%, when gap large, can timely 0.7 s, which effectively ensured travel safety people provide groundwork for practical application robots.