The Stochastic Time Delay Model and Prediction for Space Teleoperation

Space teleoperation means an extension of human operator’s intelligence to the remote telerobot in the space, and has been widely applied in space exploration and service. The main difficulty in teleoperation is the large stochastic time delay between operator and telerobot. Previous works on space teleoperation mainly focus on the control design, and some researches also give good solutions to the prediction of net work time delay, but few has been put forward to give a sufficient analysis and precise prediction for time delay in space situation. This paper conducts a study upon the time delay measure, model and prediction for space teleoperation. Time delays produced by different nodes of the spaceground round trip are measured and analyzed using statistical methods. And probability density distribution models are proposed and verified by parameter Maximum Likelihood Estimation (MLE) and χ Goodness of Fit Principle Test (GFT). Steps above show that in our teleoperation system, uplink time delay is a short-term autocorrelated sequence varying between 1.05s and 1.20s. Then, based on Non-Gaussian Auto-Regressive Model (NGAR) and Kalman Filtering (KF), uplink time delay sequence is predicted with absolute error less than 80ms and relative error below 8%. Finally, computer simulation results show that motion precision simulated by the virtual model of the telerobot will be improved with predicted value of the uplink time delay compensated, and future work is outlined in the end of this paper.