17 October 2024 in IEEE Transactions on Aerospace and Electronic Systems
This study focuses on the online thrust-loss state identification and trajectory optimization problem during the asteroid probe powered descent process where thrust-loss fault occurs. First, according to the probe's thruster configuration structure, a powered descent dynamic model of the probe and a trajectory optimization model with thrust-loss constraints are established. In addition, this paper proposes an online thrust-loss identification model based on the moving horizon estimation method. By combining the results of online thrust-loss identification, the trajectory replanning problem considering thrust-loss constraints is researched on the Bezier shape-based method, which can quickly generate feasible and relatively optimal trajectories. Finally, numerical simulations and analyses are performed to verify the effectiveness of the thrustloss identification model and the probe powered descent trajectory replanning strategy proposed in this paper. The proposed methods can ensure the soft-landing of the probe on a predetermined target, even if a thrust-loss fault occurs during the probe powered descent process
29 September 2023 in IEEE Transactions on Aerospace and Electronic Systems
A fast costate estimation technique with shaping-based method is applied to the indirect optimization of fuel-optimal multisatellite formation reconfiguration. The Bezier-curve-based shaping approach is able to solve the approximate optimal trajectory rapidly, which is to estimate the costate by being discretized at Gauss points and substituted into the discrete first-order optimality condition. The performance of costate estimation is verified in three-satellite formation reconfiguration and six-satellite formation configuration initialization. Compared with the direct method's results, the costate estimation can lead to optimal solutions of indirect optimization and only spend 5.8% and 5.9% calculation time of direct method.