Shared Control Of Teleoperation Rendezvous Based On The Consideration Of Safety

Space rendezvous is a crucial technology for complicated space missions, such as space assembly, space transpo rtation, on-orbit service and deep space exploration. For our country, there is a manual control rende zvo us technology for the backup of manned spacecraft‘s autonomous rendezvous currently; however, for unmanned spacecraft, the manual control rendezvous technology is no longer applicable. Hence, another form of backup for autonomous rendezvous is required. The research of teleoperation rendezvous is just carried out to solve this prob lem. In this dissertation, the state predictive algorithm, human handling qualities, dynamic safety evaluation model and human-machine collaboration shared control are studied. The main results achieved are summarized a s follows.A state predictive algo rithm bas ed on the principle of Smith predictor is propos ed for the teleoperation rende zvous. 1) The calculation mode l of time de lay is built, a nd the influe nces of time de lay on the stability and the performance of the control system are analyzed. 2) A time-delay compensator is designed to transfer the variable time de lay into a constant. 3) The predictive model is built on the basis of Smith predictor, and the filter algorithm is used to correct the predictive states. Expe riments are carried out based on a teleope ration rende zvous simulation system, and the results show that the predictive algor ithm has successfully reduced the effects of the time delay and that the control quality can be improved.The human handling qualities in teleoperation rende zvous are studied. 1) Based on the characteristics of relative motion and human information processing, the human operation strategy in teleoperation rendezvous is designed. 2) O n the basis of the human operation strategy, the human control model is established to emulate the human operators’ performance. 3) Experiments by the human and human control model are carried out respective ly to study the handling qualities of human operators. The results show that the operation strategy can improve the control quality of teleoperation rendezvous, the human control model can emulate human operators’ performance effectively, and the characteristics of human handling achieved by this model are accurate and believable.A real-time dynamic safety eval uation method for teleoperat ion rende zvo us is propos ed. 1) With consideration of safety influence factors, the safety corridor of teleope ration rende zvous is established, and the safety rules are designed. 2) The predictive errors propagation model is deduced considering navigation errors and control errors. 3) Based on the relative motion of the predictive states and the safety corridor, the critical safety conditions are defined, and the real-time dynamic safety evaluation model is built. Experiments based on the teleoperation rendezvous system are carried out. The results show that the predictive errors propagation model can evaluate the predictive errors accurately and that the safety evaluation mode l can grasp the safety situation well.A human-mac hine collaboration share d control algo rithm for teleoperation rendezvo us is propos ed. 1) The control modes for teleoperation rendezvous are designed, the control structure of shared control is built based on the safety evaluation model, and an artificial potential function is proposed based on the artificial potential field. 2) The conversion model of thrust is built, and the stagnation behavior is analyzed. 3) The shared weight model is established based on the principle of nine-po int controller. Experiments are carried out to testify the proposed algorithm. The results show that the human- machine collaboration shared control algorithm can improve the safety of teleope ration rende zvous not ably, and can achieve a high performance control.This dissertation studied on the safety control problem of teleoperation rendezvous, the predictive algorithm is proposed, the human handling qualities are analyzed, the safety evaluation model is established, and the human- machine collaboration shared control algorithm is studied. The achievements in this dissertation are valuable, and can be used as the reference for the engineering application of teleoperation rende zvous in the future.