Research On Control System Of Rescue Robot And Information Process

The disaster environment is so complicated after Earthquake, Tsunami, Fire and War that rescue work can be hardly done. Meanwhile, because of its dangerous, lives of these rescue teams as well as masses of victims are under great threats. However, the Rescue Robots can help even replace these rescue workers to save lives in extremely dangerous situation, and in this way, the efficiency of rescue work can be greatly raised. Therefore, the research and exploration of Rescue Robots have become one of the most important fields in recent practical application of robot technology.Aiming at Tracked Rescue Robots that take part in the RoboCup Rescue competition,this thesis designs a kind of rescue robots control system whose control center is based on the Freescale MC9S12XS128 microcontroller and does some research on the procession of information. All these researches in my thesis all have taken practical application into consideration. As for the researches in the thesis, on one hand, the control system mainly aims at the hardware platform, and develops corresponding software, by doing so, the attitude control is accomplished. On the other hand, laser radar data used in environmental map building is analyzed in the aspect of information procession, and an environmental map is efficiently set up by practical experiments. Therefore,the algorithm's validity and rationality are proved. The main work of the thesis and achievement are summed up as follows:First of all: to confirm that the Tracked Rescue Robot is the carrier based on the needs of the competition; to pick out the models of drive motors according to robots' basic demands,which include the main drive motor, main arm motor, and auxiliary arm motor.Secondly: considering the rescue robots' functional needs, characteristics of the motion system and practical analysis on the control system, a general scheme is determined, that is PC-based position machine and the next-bit machine distributed control system. The position machine is responsible for information fusion, environmental map building and video data receiving and display. While the next-bit machine is for the laser radar and data acquisition and transmitting (i.e.: MTi data). Because the position machine and the next-bit machine have clear duty and work range, the processing speed of the control system and the information processing system is raised. And in this way,the stability and real-time of the robot system are guaranteed.Based on the general scheme of control system and information procession system, the hardware platform of the control system and information procession system is designed in this thesis, including the designation of bottom control circuit, video communication system and map building system, whose control centers are all Freescale MC9S12XS128.Moreover, the designation of control system, the designation of interface program of information procession system and the designation of data packing and sending program are accomplished on the foundation of hardware; among these designations mentioned above, the diagram software is carried out specifically, which divides into two parts: the position machine and the next-bit machine. The next-bit machine collects laser radar data through serial port, and then sends data to the position machine by wireless network. Meanwhile, once the position machine received these laser radar data, it will perform the program of building map, and achieve the environmental map.Finally, the research of this thesis is focused on the laser radar data applied in environmental map building. In the first part of this section, the principle of the application of laser radar data in environmental map building is described,and the effect of the experimental map building when there has existing error of sensor is presented as well. in the second part, a new environmental map building that bases on linear feature is presented because of the existing error, which divides the environmental map building to two steps: the feature extraction and the feature matching. The feature extraction adopts an improved Split-and-Merge algorithm and extracts linear feature of data point set by laser radar. On the other hand, the feature matching is divided into the steering motion; so in this section, the linear motion and the ways of motion of the two kinds of robots are under discussion, and presents the algorithm of the feature matching. At last, the thesis introduces the procession of the diagram algorithm, and presents the practical building effect of the algorithm.