Explore the vast range of titles available.

The lunar probe often has some remaining fuel on completing the predefined Moon exploration mission and may carry out some additional tasks from the Moon orbit using the fuel. The possibility for the lunar probe to escape from the Moon and the Earth is analyzed. Design and optimization of the trajectory from the Moon orbit to the Near Earth Asteroids （NEAs） using the spacecraft＇s residual fuel is studied. At first, the semi-major axis, inclinations and the phase relations with the Earth of all the numbered NEAs are investigated to preliminarily select the possible targets. Based on the Sun-centered two-body problem, the launch window and the asteroid candidates are determined by calculating the minimum delta-v for two-impulse rendezvous mission and one-impulse flyby mission, respectively. For a precise designed trajectory, a full ephemeris dynamical model, which includes gravities of the Sun, the planets and the Moon, is adopted by reading the JPL ephemeris. The departure time, arrival time, burning time duration and thrust angles are set as variables to be designed and optimized. The optimization problem is solved via the Particle Swarm Optimization （PSO） algorithm. Moreover, two feasible NEA flyby missions are presented.

The Unified S-Band （USB） ranging/Doppler system and the Very Long Baseline Interferometry （VLBI） system as the ground tracking system jointly supported the lunar orbit capture of both Chang＇E-2 （CE-2） and Chang＇E-1 （CE-1） missions. The tracking system is also responsible for providing precise orbits for scientific data processing. New VLBI equipment and data processing strategies have been proposed based on CE-1 experiences and implemented for CE-2. In this work the role VLBI tracking data played was reassessed through precision orbit determination （POD） experiments for CE-2. Significant improve- ment in terms of both VLBI delay and delay rate data accuracy was achieved with the noise level of X-band band-width syn- thesis delay data reaching 0.2-0.3 ns. Short-arc orbit determination experiments showed that the combination of only 15 min＇s range and VLBI data was able to improve the accuracy of 3 h＇s orbit using range data only by a 1-1.5 order of magnitude, confirming a similar conclusion for CE-1. Moreover, because of the accuracy improvement, VLBI data was able to contribute to CE-2＇s long-arc POD especially in the along-track and orbital normal directions. Orbital accuracy was assessed through the orbital overlapping analysis （2 h arc overlapping for 18 h POD arc）. Compared with about 100 m position error of CE-l＇s 200 kin x 200 km lunar orbit, for CE-2＇s 100 km x 100 km lunar orbit, the position errors were better than 31 and 6 m in the radial direction, and for CE-2＇s 15 km~100 km orbit, the position errors were better than 45 and 12 m in the radial direction. In addi- tion, in trying to analyze the Delta Differential One-Way Ranging （ADOR） experiments data we concluded that the accuracy of ADOR delay was dramatically improved with the noise level better than 0.1 ns and systematic errors better calibrated, and the Short-arc POD tests with ADOR data showed excellent results. Although unable to support the development of an independent lunar gravity model, the tracking data of CE-2 provided evaluations of different lunar gravity models through POD. It is found that for the 100 km x 100 km lunar orbit, with a degree and order expansion up to 165, JPL＇s gravity model LP165P did not show noticeable improvement over Japan＇s SGM series models （100x100）, but for the 15 kmxl00 km lunar orbit, a higher de- gree-order model can significantly improve the orbit accuracy.