Research On Robust Trajectory Planning For Non-Cooperative Target Rendezvous With Constraints

Considering the limited communication and observation for non-cooperative target rendezvous with uncertainties and multi-constraints,the robust,intelligent,autonomous,safe and reliable rendezvous is of great importance in both theory and application.Aiming at the requirements of the on-orbit service,this paper studies the robust trajectory planning problem for non-cooperative target rendezvous with uncertainties and multiconstraints.The robust and intelligent trajectory planning and control methods for perturbed elliptic orbit rendezvous with uncertainties and multi-constraints are based on second-order cone programming,state transition tensor theory,neural network and preset performance control.Main innovative works are as follows:(1)Fast rendezvous trajectory planning for space non-cooperative target.A fast rendezvous trajectory planning method for elliptical orbit rendezvous with multi-constraints based on second-order cone programming is proposed.First,a relative motion model considering J2 perturbation,atmospheric drag perturbation,and orbit eccentricity.Second,an optimization problem with keep-out zone,approach corridor,state constraints and objective function is established.Finally,the optimization problem is transformed into and solved by standard second-order cone programming by statically-attached planes and covariate relaxation technique.(2)Uncertainty analysis for non-cooperative target rendezvous.First,uncertainty model including environmental error,sensor measurement error,actuator error and initial state error is established.Second,the uncertain propagation,update and correction equations of closed-loop GN&C system are derived based on the state transition tensor theory.Finally,uncertainty and error budget curve for non-cooperative target rendezvous are analyzed.(3)Optimal selection of sensors and actuators for non-cooperative target rendezvous.First,uncertainty propagation model of closed-loop GN&C system with sensor error and execution maneuver error is established using uncertainty analysis.Second,the optimal selection of sensors and actuators problem is modeled by Kronecker theory and successive approximation theory.Finally,optimal selection of sensors and actuators problem is solved by second-order cone programming.(4)Robust trajectory planning for non-cooperative target rendezvous with multiconstraints.First,a robust trajectory planning problem is established using perturbed elliptical relative motion model and uncertainty propagation model.Second a objective function including expected control consumption and uncertain control consumption and the terminal uncertainty constraint are introduced into the robust trajectory planning problem.Finally,the robust trajectory planning problem is solved by a genetic algorithm.(5)Final trajectory control for non-cooperative target rendezvous.Aiming at eliminating the impact of uncertainties and actuator failure in the final approach for noncooperative target rendezvous,a trajectory control method is proposed.First,the uncertainties and actuator failure is detected by long-short term memory.Then,the prescribed performance control is used to control the trajectory of with prescribed transient and steady performance,achieving high precision rendezvous and docking.