Research On The System Of Polishing For Industrial Robot Through Force-feedback

It's common in the automated process that manipulators executes a task involving interaction with the environment,typical examples include polishing,deburring,machining and assembly,etc.The compliance of manipulators is crucial for the performance of those industrial tasks.Active compliance control of manipulator is an another important research field of robot control,which is different from the pure position control.To improve both the compliance and the efficiency of those automated tasks,manipulator control system based on force-feedback is designed,a method for processing and calibrating the sensing information is proposed,and a robot trajectory control method and force tracking method is presented based on on-line control strategy.A open-architecture robot control system integrated with force sensing system is designed and implemented.And the overall scheme of the control system based on “IPC(Industrial Personal Computer)+motion controller”is expounded.An software for the active compliance control system is developed and the force data transmission between main host computer and the robot controller is devised,and the operation flow of the software system is designed.By analyzing the noise of the environment,the Kalman filter is used to filter the identifying force signals(IFS)on line.The mathematical model of the IFS is established and the filtering experiments are conducted both in the cases of stationary and non-stationary signal.The experiment results show that the noise is filtered and the precision of IFS is obviously improved.Meanwhile,by considering the zero drift of 6-axis force/torque sensor and the angle bias of robot installation,we proposes a method for estimating and compensating the load and the sensor bias.Several experiments are conducted and the results show that the proposed method satisfies most of the industrial tasks' need.In order to achieve an contact control of robot to the environment,the environment spring equivalence value is estimated by BP neural network.By analyzing the model of the robot and the environment,the adjustment strategy of the trajectory direction of robot is presented and the online adjustment algorithm is designed.Combining with the PID controller,the trajectory direction of robot is corrected on-line through the proposed adjustment algorithm.The experiments result shows that the robot achieves a stable trajectory with the environment.The impedance control model between the displacement of robot trajectory and the deviation of interaction force is established.The position-based constant impedance control strategy,which is implemented in a discretization form and combined with the direction adjustment algorithm,is designed to achieve better force tracking performance.The steps of achieve the on-line force tracking tasks is summarized,and several experiments using the designed strategy in two kinds of work piece are conducted.The results show that the average of force tracking error is within 5N.