Research On Key Techniques Of Arm Target Capture Based On Electric Wheelchair

Nowadays,the aging of the population is increasing,so it is particularly important to find a solution to reduce this pressure on society.How to improve the free movement space of the elderly and the disabled has become a hot topic.The rise of intelligence has caused an upsurge in the research of serviee robots.Intelligent wheelchairs which eontains manipulator have the characteristics of wheelchair movement and manipulator flexibility.Therefore,it can play an auxiliary role in the lives of the elderly and disabled people and can expand their freedom of life.However,most of the control methods of the device are manual operation,which still bring many problems to the user.Therefore,it is necessary to study its intelligent control.In this paper,the research mainly includes the location of target and the control of manipulator.Target position is achieved through image processing methods,including image acquisition,image preprocessing,image segmentation.The manipulator control mainly includes the establishment of the manipulator space coordinate system,kinematics analysis,the manipulator working space description and the end of manipulator trajectory planning.Wheelchair as a mobile platform,the motor driving mode is briefly introduced and drive circuit is designed.In order to realize the target position,the image is preprocessed by the bilateral filter,and the color space conversion,so the target can be easier separated from the background.Using the H and S components of the HSI as growth criteria,similar pixels are grown into the same region by using the growth area algorithm.Because of the PCNN synchronous pulse emission phenomenon,the PCNN algorithm finally achieved the image segmentation and obtained the pixel coordinates of the target position.Through the target recognition experiments,it is known that this method can effectively avoid the phenomenon that the traditional PCNN segmentation method can not be divided correctly due to the influence of light factors,and this method has certain validity.Aiming at the problem of loss depth information of monocular vision,a method of monocular vision ranging is proposed,which is proved to be feasible by experiment.Using the D-H parameter method,a spatial coordinate system is established for the six-axis manipulator,and a kinematics model of the manipulator is established.On the basis of the model,the positive kinematics equation and the inverse kinematics equation can be deduced,and the relationship between the rotation angle of the rotating axis of each arm and the position of the end of the manipulator can be obtained.The kinematics equation is verified by the Robotics toolbox and the manipulator trajectory and workspace are also simulated and analyzed by it.The results prove the correctness of the research on the manipulator control.Finally,the manipulator control interface is designed to realize human-machine interaction function.