I do research on the moment based visual servoing . Due to nonlinear nature of the system, proportional integral sliding mode controller is capable to improve the performance and robustness of visual servoing. This method increases the domain of attraction even in extraneous image regions, and furthermore, it achieves convergence to the target point despite of uncertainties. Different set of experiments are performed on 5 DOF industrial robot with one degree redundancy in order to examine the performance of the system. The system is equipped with the wide lens camera on the end-effector.
Setting suitable gain and finding appropriate solution for the aforementioned method needs a lot of tests. In order to ease the process of testing and decrease the risk of robot failure, a simulator has been developed in ROS and Gazebo environment by cooperation of ARAS visual servo team. In the presented simulator objects consider to be planar and 4 DOF camera motion are controlled by image moments. Complete code of the simulator can be accessed through Github site. (aras_visual_servo)
