Research On Enhancing The Quality Of Remote Communications For Mobile Robot

Regarding automobile robot, its remote communication is a fundamental capability. However, the low capability of the present network performance is in contradiction with a great deal of communication datum. This makes a great impediment for us to design a stable communication system. The main content for this paper is to make a research on how to enhance the quality of the remote communication in view of the performance requirement of the robot's multi-sensors data.The paper introduces the multi-sensors fused remote communication system, proposing RTP and the TCP protocol union use strategy regarding the multi-sensors data communication. Based on in-depth research on RTP/RTCP protocol and the congestion of network, a systematic research on the dynamic monitoring and control policies for the quality of video is conducted, the formulas for calculation the monitoring parameters of QoS are inferred, and the policies for dynamic control of QoS are proposed. A new scheme of congestion control which is applicable to a synthetical network composed of wired and unwired is conducted. The paper also improves upon the conventional auto-adapted arithmetic, so as to reduce the loss rate of RTP passages, at the same time enhance the use factor of the network.According to practical application, this paper also thoroughly studies neural net in the aspect of distributing resource of network, designing a BP model used to predict the use factor of bandwidth. Thus, we resolve the big delay phenomena of TCP data stream in the process of the communication.This research on enhancing the quality of the remote communications of the automobile robot provides the condition to robot's monitoring and controlling, the sensor survey data processing as well as the real-time diagnosing sensors.