Autonomous Low‐Thrust Guidance: Application to SMART‐1 and BepiColombo

Abstract

Several techniques have been developed to obtain optimum trajectories with low-thrust propulsion. However, few low-thrust guidance schemes have been investigated to fly the reference optimum trajectories. The guidance algorithm successfully employed in the DeepSpace1 mission was the first approximation through the presented guidance schemes, valid for various interplanetary low-thrust trajectories, independently of the optimization technique they result from. A method is presented to transform any given thrust profile to a thrust law defined by a finite set of control variables. This law allows the definition of a control vector to be optimized for the guidance purposes. Simulations were carried out to compare the performances of the algorithms to very different missions, such as SMART-1 and BepiColombo. The good performance of the enhanced guidance schemes prove the generic applicability of the algorithm. Parametric analysis allows the assessment of stability and robustness of the schemes and the sensitivity to certain parameters. Table.

Low-thrust trajectory design as a constrained global optimization problem

“…Here, the spacecraft leaves Earth on 13 April 2010, performs two flybys at Venus to lower its orbit, then encounters Mercury twice to lower its V 1 to 0.5 km/s before it arrives on 5 September 2015. Comparing such trajectory with the one described in the BepiColombo early mission planning stage, it is noted that the results are comparable with those described in [33] (as cited in [34]) with regard to the overall shape of the optimal trajectory, which includes the planetary resonances, the times of flight, and the engine thrust periods, but have been obtained using a completely automated process.…”

Optimization of low-thrust many-revolution transfers and Lyapunov-based guidance

“…Moreover, the mean costate variables at the initial time are needed for trajectory optimization. In this study, we considered that the control law is directly governed byk = [¯ p¯ f¯ g¯ h¯ k ] T , and the time history ofk(t) is interpolated through a finite number of nodal values along the time axis instead of governed by the mean costate differential equation (9). The nodal values for interpolatingk(t) should be optimally chosen.…”

“…Kluever [8] developed a well-performed tracking guidance using an inverse dynamics approach. Gil-Fernandez et al [9] presented a tracking guidance for Smart-1 and BepiColombo interplanetary missions. Another category of the low-thrust feedback guidance is nonlinear feedback control without tracking reference trajectories.…”

Optimization and guidance of very low-thrust transfers to geostationary orbit

“…Therefore, once the optimal free-phase (or longitudeunconstrained) trajectory is found, the coordinates of the nodes are slightly changed to fulfill the phasing constraint while minimizing the variation of the flight time. The constrained parameter optimization is described in [5]. It is important to note that in this case the evaluation of the constraint involves the optimal control solution of the fast problem.…”