Abstarct: Due to the growth of knowledge in the field of robotics, certain application requires robots to be more accurate than ever. To get and keep the flexible and rigid bodies with grippers, different clamp design with different shapes and with many methods has been controlled. object can vary in size, shape and behavior. Therefore precise method of control is needed to identify the nature of the object and on the issue control commands. Robotic grippers need a centralized system to control the flow of data from sensors and then calculate direct visual contact and how to perform a gripping procedure. Naturally human shortcut the gripping procedure and select the shortest path in order to simplify the task. Traditional control methods are being taken to secure the object. In summary, previous research on the modeling and control of robots receiver is done by Anfis main research topics are pursued the path of the robot, and the chain. The application of Anfis in the body by paw received less attention, although research is ongoing to secure the body to the conventional control methods. To innovate and achieve a new method, fuzzy rules for identifying the behavior of the object in the Anfis activation functions have been used in Anfis network. For this purpose, a set of basic data on the behavior of the object and the gripper is used to model the force-position control of the gripper. The basic part of this project is to develop a model that treats the object as well the neuro-fuzzy model had detected. It should also control the claw to the body in order to not add extra force into the subtle objects, causing damage to them.