Design of real-time transmission system combined with object detection algorithm

Abstract

With the rapid development of robotics and artificial intelligence technology, robots are increasingly appearing in the rescue scene of natural disasters, terrorist activities and various types of unexpected accidents. If the rescue robot wants to complete the task quickly and accurately, the command personnel are crucial to the remote control of the robot. As the "eye" of the robot, image transmission technology is the key to realize the remote control of the robot. Image transmission is a technology that transmits the image from the robot's perspective to the operator in real time to assist the operator to achieve remote control. With the increasing complexity of the work involved in robots, the traditional real-time image transmission system is gradually difficult to meet the needs.In the traditional robot control system, there is a completely separate relationship between the acquisition of visual information and the control of motion, which leads to the inevitable hysteresis of motion control. In order to meet the new requirements of the robot image transmission system, this paper presents a design scheme of real-time image transmission system combined with object detection algorithm. The target coordinates identified by the algorithm are directly sent back to the motion control mechanism, and the visual information and motion control are directly connected to improve the overall response speed of the system. The specific implementation is to obtain the image data through ESP32, and then send it to the host through the network. The host realizes the target detection, and then sends the detection results back to ESP32 through the cloud server, and ESP32 transmits messages to the main control of the motion system. The design scheme has been tested in practice and has good recognition effect and stability.