The Research On Coordinated And Tracking Control Of The Two-wheeled Self-balancing Vehicle Based On Machine Vision

As a typical representative of the two-wheeled self-balancing robot, the two-wheeled self-balancing vehicle is one of the typical research object of the imitation of Human Intelligent Robotics. But it was forgotten by people since its creation in1986 gradually, until the rise of “Segway” in this years, its powerful and practical significance in military and civilian field has been found by more and more people. And it attracts more and more domestic and foreign researchers join the research ranks. At present the mostly domestic and foreign research about it, focus on the study of individual behavior and few researchers study its multiple individual behavior, such as coordination, tracking, imitation and cooperation etc. Therefore, this paper takes two sets of two-wheeled self-balancing vehicle as the research system and attempts to achieve the coordinated tracking in this system based on machine vision. In addition,the collaborative tracking is popular the literal meaning understanding, that following the car along the walk in a straight line to the target vehicle line tracking and collaborative travels in straight line which different from the meaning of the domain of more intelligent systems refers to the controller design only by Bureau of the Ministry of information instead of global information.In the absence of domestic and foreign research data, in order to guide the research work in this paper summarize the development history of the two-wheeled self-balancing and the mainly control methods of the four-wheeled intelligent vehicle. And the research work of this paper is mainly divided into four parts. The first part is the research of the system internal model. At this part, Firstly, this paper deduces the two-wheeled self-balancing system dynamics model and the state-space model. Secondly, this paper mainly analysis the multi-body volatility model when its in dynamic balancing. Thirdly, this paper deduces the vehicle distance when the two-wheeled self-balancing vehicles in coordinated tracking process of moving. Fourthly, this paper analysis the volatility model influence on the image identification, the process of image collected and the vehicle distance and vehicle speed.The second part is the image theory research and the simulation analysis about the machine vision. At this part, firstly, this paper discusses the image enhancement and processing method, such as the image gray level correction, the different smoothing filtering method and the treatment effect. Secondly, this paper introduces the description of the image effective region, image distance and image shape. Thirdly, this paper analysis the Matlab simulation, through using Matlab simulation analysis to determine the threshold value of image binaryzation, the coefficient value of linear transformation, the result of smoothing filtering, the coordinates of image effective region, the image shape parameter range and vehicle distance range.The third part is the control method design and the Matlab simulation experimental results of the system. Firstly, this paper analysis the system zero-input response when the body angle as output. Secondly, this paper analysis the effect of PID controller through using the Matlab simulation when the body angle and wheel speed as output. Thirdly, this paper design the control strategy of the vehicle distance, speed and direction. Fourthly, this paper introduces the Matlab simulation results of the vehicle tracking distance and speed.The fourth part is the introduction of the system platform and the experimental results. Firstly, this paper introduces the whole experimental environment. Secondly, this paper introduces the hardware environment and the software program design. Thirdly, this paper introduces and summarize the experimental results.