Research On Image-Based Uncalibrated Visual Servo Robotic System

Visual servoing has been one of the most active topics in robotics.It involves many disciplines such as manipulator-arm kinematics,dynamics,and joint-level control,as well as camera modeling,image processing,feature extraction,it is a kind of integrated research.Uncalibrated visual servoing system reveals preferable flexibility and adaptability in comparison with the classical visual servo systems which require system calibration,therefore it becomes a significant branch and hotspot of research in the field of visual servoing.In the first two chapters,this thesis reviews the researc progress of uncalibrated visual servo control system in recent years from two perspectives,which are controller design and motion planning.Then the Denavit-Hartenberg notation is used to describe the geometry of a serial chain of links and joints.We can compute the relative pose of each link as a function of its joint variable and compose these into the pose of the robot's end-effector relative to its base.Next,a new adaptive controller is presented in this thesis for image-based dynamic control of a robot manipulator using a fixed camera who's intrinsic and extrinsic parameters are not known.This adaptive algorithm combines the depth-independent interaction matrix,Slotine-Li method with on-line minimization of the errors between the real and estimated projections of the feature points on the image plane.Finally,a new adaptive controller is presented in this thesis for image-based tracking of a robot manipulator without using visual velocity when the intrinsic and extrinsic parameters of the camera are not calibrated.The visual velocity is estimated by the joint velocity of the robot manipulator through the image Jacobian matrix.An adaptive algorithm is used to estimate the unknown parameters on-line.Then,the asymptotic convergence of the tracking errors on the image plane is rigorously proved by the Lyapunov theory with the full consideration of the nonlinear dynamics.