Research On Bionic Eye Movement Control Method

After thousands of years of evolution,human beings have formed a fully functional visual system,which shows unique advantages in fast and accurate tracking targets,perception and recognition processing,and rapid switching fixation points.The appearance of machine vision makes the machine have the ability to interact with the external environment,but it has obvious limitations in terms of accurate perception,motion tracking,rapid recognition processing and so on.Therefore,it is of great significance to study the neurophysiological mechanisms of the human eye movement deeply.After that,we can construct bionic eye similar with human eyes'function from the perspective of bionics by combining with movement control algorithms and computer vision techniques.This paper concentrates on studying the bionic eye movement control algorithm,and the main work can be concluded as follows:Firstly,we analyze the research background and significance of the bionic eye,review and summarize the current research work of conjugate eye movements,disjunctive eye movement,reflex eye movements and multiple eye movements,point out the characteristics and the main problems of them in software and hardware,and introduce the main work and chapter arrangement of this paper.Secondly,we study the eye movements from the point of neurophysiology.The structure of the eyeballs and the function of each part of them are studied,and the movement mechanism that the extraocular muscle drives eyeballs moving is analyzed.We mainly introduce parameter index and neural control mechanism of conjugate eye movements,disjunctive eye movement and reflex eye movements which provides theoretical support for building the eye movement control models.Finally,the visual pathway of human beings is studied.Thirdly,we study the neural control mechanism of the smooth pursuit movement of mammals and propose an information fusion control with time delay for smooth pursuit.By fusing the soft constraint information of the target trajectory of eyes and the ideal control strategy,and the hard constraint information of the eye system state equation and the output equation,optimal estimations of the control variable are obtained according to the theory of information fusion estimation.The proposed control method can implement smooth pursuit movement with prediction and learning that eyes can track not only periodic moving targets but also nonperiodic as well as solving the problem of time delays in the process of image processing and signal transmission and so on.The simulation experiment results prove the effectivity of the proposed method.Fourthly,we study the neural control mechanism of the vergence eye movement,improve Erkelens's model and propose a vergence movement model with prediction and learning.The proposed model can predict and learn the current target vergence used RLS algorithm,and record the learned target motion using the internal model.The proposed method can implement vergence movement with prediction and learning that eyes can track not only static targets but also continuous moving targets.The simulation experiment results prove that the proposed method has excellent tracking performance.Finally,we analyze the characteristics and neural mechanism of the saccade-vergence eye movement and build a saccade model and vergence model respectively based on neurophysiology.Then we study the interaction between saccade and vergence movement and build a model for saccade-vergence interaction.The model combines this two eye movements by saccade-related vergence burst neurons(SVBN)that can make eyes track targets in three-dimensional space.The simulation results consist with the characteristics of the saccade-vergence movement and the velocity of eyes satisfies the displacement and direction of eye movement.