Stabilization and control of teleoperation systems with time delays

Abstract

Time delayed teleoperation has been one of the first and most challenging topics in robotic control. Although numerous methods have been developed by researchers all over the world during the last two decades, those methods have limitations or need special assumptions to be applied. With the development of the world-wide-web, teleoperation through the Internet sees a bright future. However, the constantly changing time delay in Internet data transmission brings a big challenge to Internet-based teleoperation. These time delays not only degrade the system performance, they also destabilize the teleoperation systems. In this work, a control scheme for teleoperation systems with time delay is developed based on the concept of passivity. This control method requires neither detailed knowledge of the manipulator systems nor the mathematical models of the environments, and it is applicable for any time delays. The model independence and time delay independence make the proposed control method well suited for teleoperation in the real world, which includes remote site explorations, tele-surgery, space explorations, and teleoperation through the Internet. The main contribution of this method is that it is less conservative than the traditional passivity based method. In our method, the passivity controller only operates when the system loses passivity, while in a traditional passivity formulation, the controller works at all times during operation and thus adversely affect the performance of the system. Using the proposed control scheme, a sub-system is defined that is composed of the communication channel, slave robot and the manipulated environment. This sub system is treated as a one-port network component, and passivity theory is applied to this component to assure stability. The energy flowing into the one-port network, in the form of the control command and the force feedback, is monitored. The passivity condition is violated when the net inflow of energy becomes negative, indicating that this component starts to generate energy, causing system instability. To reinstate the passivity of the network, a passivity regulator is activated to modify the feedback force to the master, and thus adjust the energy exchange between the master and the communication channel. Using the passivity regulator, the passivity of the system is maintained. When this method is applied, only the information at the interface between the master manipulator and the communication channel is collected and observed, there is no need for accurate or detailed knowledge of the structure or timing of the communication channel. The method can make the system lossless regardless of the feedback force, the coordinating force controlling the slave joint motions, or the contact force. The approach presented here can stabilize the system regardless of the time delay, discontinuities with environmental contact, or discretization of the physical plant. Thus, unlike the traditional method, it will pose no problem when the environmental contact force is directly fed back rather than the coordinating force controlling the slave robot motion. The results of this work show that it is advantageous to use the measured environmental force as the feedback, providing superior performance for free motion and more realistic haptic feedback for the operator from the remote environment. Along with computer simulations of a generic master-slave teleoperation configuration, experimental results are presented to validated the simulations and verify the proposed control scheme. A Microsoft Sidewinder force feedback Pro Joystick has been used as master and a PUMA 560 robot has been used as slave. The experimental results show that ix the proposed method can stabilize the teleoperation system with any time delays and with any working environment. Both simulations and experiments show good position and force following performance.