Vision-assisted robot kinematic error estimation: Principle and application

Vision-assisted robot kinematic error estimation is a robot vision application to find out "where it is" of the robot or the surrounding objects. The vision system demonstrates the observation of the robot motions in the image form, therefore it can perform the kinematic error estimation in the image feature space and then transform it into 3D robot space or camera space if necessary. The interrelated theories for the vision-assisted kinematic error estimation are discussed in this thesis.; A direct visual servo control system is an application of robot vision-assisted kinematic error estimation. The robot joint displacement errors are measured in the image feature space by the vision system and sent to the joint controller. As a case study, a direct visual servo control method for a SCARA type manipulator is discussed and the corresponding simulations are presented in this thesis.; The direct visual servo control method could be an alternative control scheme when one or more internal joint sensors fail. The possible functions of a vision system in the three-level (servo level, interface level and supervisor level) fault tolerant system of a robot are presented. Simulations demonstrated that a direct visual servo control system could be an alternative controller to fulfill full recovery and partial recovery scheme at servo level. This application shows a potential method to improve the fault tolerance capability of an existing robot by employing a vision system.