Research Of Positioning And Tracking System For Intelligent Wheelchair Based On Monocular Vision

With the increasing aging of China’s population,the society is paying more and more attention to the elderly and the disabled,which has led to an increasing demand for wheelchairs.Highly intelligent and autonomous wheelchair types have become a research hotspot in the field of intelligent service robots.The smart wheelchairs currently on the market have complex structures and are expensive.Ordinary families cannot afford smart wheelchairs,so it is of great research significance to design a low-cost smart wheelchair suitable for the general public.Visual recognition is the key to the field of intelligent wheelchair recognition.Due to the complex outdoor environment,the large amount of image data,and the timeconsuming processing,it is difficult to meet the real-time performance and robustness.At the same time,in order to meet the accuracy of target recognition and reduce the influence of camera distortion,expensive cameras are usually used.In this paper,a low-cost monocular camera and an Open CV-based color center localization method are used to propose a new intelligent wheelchair tracking method.And established a mathematical model of intelligent wheelchair positioning and tracking.The parameters of the mathematical model are determined through experiments and calculations.Through theoretical derivation and a large number of experimental tests,the correctness,feasibility and effectiveness of the method are verified.The system resolution is 2mm,and the data processing speed reaches 30 frames per second.The main research contents of this paper are as follows:(1)According to the structure of the intelligent wheelchair and the actual needs of users,a new,inexpensive and practical target recognition and tracking method is proposed.It is mainly based on the color center positioning function of Open CV,combined with the principle of the camera vision model and the current software and hardware.And the method has been proven to be feasible.(2)A suitable camera and in-wheel motor have been investigated in the market,and an efficient visualization software been selected.Through the calculation and functional requirement analysis of the mechanical structure of the wheelchair,the size and rotational speed of the in-wheel motor,the wheelchair can realize a variety of functional modes,and the kinematic equations are established.By Open CV vision technology,the influence of illumination,noise,etc.can be eliminated or reduced,and then the effective information in the image can be extracted by methods such as color model conversion,image smoothing,graphics segmentation,and centroid calculation,so as to realize the software programming of target recognition..(3)Through a large number of experiments,it is verified that the programming program can accurately identify the target,and the principle accuracy and antiinterference ability based on this method are verified,and the experimental results are mathematically analyzed.At the same time,the influence of the color and shape of the target on the measurement system is experimentally studied,and the target type of the system is selected.(4)Different kinds of mathematical models are established and compared.And the fitting equation with the highest precision is selected as the mathematical model of the system.The final target recognition and positioning system is completed,and the experimental results of the system are analyzed to obtain the accuracy of the system.(5)The UI interface of the intelligent wheelchair system is designed based on QT,which can display and mark the position of the target center of mass and the actual distance.