Research And Implementation Of Networked Control And Automatic Obstacle Avoidance For Mobile Robot

The networked control of robot means that the computer controls the remote robot through networks and is equipped with wide application in future. With the environment increasingly complex for application, the robot itself is required to possess some certain abilities of perception and able to deal with some unexpected events. So the combination of remote control of robot based on network and improvement of robot's autonomic abilities is gradually concerned by people.This paper mainly designs and realizes the mobile robot system based on the combination of network control and autonomic avoidance of obstacle. The main work of this paper is listed as following aspects.(1) In order to complete the mission for Video capture and wireless communications, robot has been extended an ARM development board on which the video capture module and wireless communication module have been designed.(2) To point against the real-time of the video capture system, we choose the capture device that support the hardware image compression and the real-time video capture method based on ARM-Linux, which greatly improve the real-time of image acquisition.(3) In order to facilitate the mobile robot to freely work in or out some environment difficult to plan net lines, we design and realize the wireless communications between server and mobile robot; To point against the real-time of the video transmission, we used the Wireless LAN access technology based on the IEEE802.1 l;In order to ensure the reliability of network transmission, we used the technology of network programming based on the C/S frame to realize the communications between server and client.(4) In order to enhance the autonomy of mobile robot, we design and realize an autonomic obstacle avoidance strategy for mobile robot with combination of single camera and Infrared ranging sensors in unknown environment. In order to obtain the distributed information of the front obstacle, we take the optical flow calculation on the acquired image sequences .At the same time, we use Infrared ranging sensors to get the distance information of the obstacle for setting up the obstacle avoidance strategy. The reliability of the strategy is better than the one only depending on the vision sensor or depending on ranging sensors, and tests the whole system through experiments.