Research Of The Small Visual Tracked Robot Based On DSP

Mobile robot is a integrated system that integrated multiple functions,including environmental perception, behavior control, dynamic decision and other functions. It has an ability of highly self-adapting, self-organizing and self-layout, which can adapt to motion in the complex condition. When mobile robot is working, the first problem need to be solved is percepting the external environment, in order to provide information for further decision-making. As the vision system can get the rich information, the mobile robot with visual function has become the trend of development. Small visual tracked mobile robot based on DSP has been related studied in the paper, with tracked mobile robot as research object and with vision system as main line of the paper.The overall structure of the tracked mobile robot is designed. Mobile robot used AVR microcontroller as the control chip, using multiple sensors as signal acquisition, and use two DC motors as the drive.The vision system of the robot is designed. The vision system adopts TI's TMS320DM642DSP chip as the core processor. Using TI's TMS320DM642DSP chip as the control core of the vision system, which can be used for a large number of digital processing, but also ensure the system's real-time performance. In addition, according to the need, the corresponding peripheral circuit modules are designed, including video decoding circuits, communication circuits, memory circuits, the reset circuit, and so on.The image processing algorithms is researched, including image enhancement, image denoising, image segmentation, image recognition, image positioning. By analyzing and comparing each algorithms, and combining with actual situation to choose the appropriate algorithms. And simulation to verify the effectiveness of the selected algorithm.Finally, the robot path planning is studied. By analyzing the traditional artificial potential field method, a robot path planning based on the improved artificial potential field method is presented, and through Matlab simulation to prove its effectiveness.