Trajectory Design for the ESA LISA Mission

Abstract

The three Laser Interferometer Space Antenna (LISA) spacecraft are going to be placed in a triangular formation in an Earth-trailing or Earth-leading orbit. They will be launched together on a single rocket and transferred to that science orbit using Solar Electric Propulsion. Since the transfer Δv depends on the chosen science orbit, both transfer and science orbit have been optimised together. For a thrust level of 90 mN, an allocation of 1092 m/s per spacecraft is sufficient for an all-year launch in 2034. For every launch month a dedicated science orbit is designed with a corner angle variation of 60° ± 1.0° and an arm length rate of maximum 10 m/s. Moreover, a detailed navigation analysis of the science orbit insertion and the impact on insertion errors on the constellation stability has been conducted. The analysis shows that Range/Doppler measurements together with a series of correction manoeuvres at the beginning of the science orbit phase can reduce insertion dispersions to a level where corner angle variations remain at about 60° ± 1.1° at 99% C.L. However, the situation can become significantly worse if the self-gravity accelerations acting during the science orbit phase are not sufficiently characterised prior to science orbit insertion.

LISAmax: improving the low-frequency gravitational-wave sensitivity by two orders of magnitude

“…The full evolution of both quantities is shown in figure 2. These variations are significantly smaller that what is achievable for LISA [21] ((60.0 ± 1.0) • for the corner angles and ±10.0 m s −1 for the arm length rates). This is due to the larger distance from perturbing sources, mainly the Earth.…”

“…As shown in figure 3(a), the resulting dispersion on these parameters is rather limited despite the conservative assumptions, owing to the quiet environment in the vicinity of the libration points. Note that for LISA the resulting disturbance is more significant [21].…”

“…For the design of the LISAmax science orbit, we assume similar requirements as for LISA [21] extending into LISAmax frequency range.…”

“…Since LISAmax is assumed to have two test masses (as LISA), the situation is more complicated in reality. For more details, refer to [21]. For the purpose of the current analysis, the relevant result is that it is only possible to predict the magnitude and direction of a part of the selfgravity-induced accelerations.…”

Stochastic gravitational wave background reconstruction for a non-equilateral and unequal-noise LISA constellation

“…The AET basis is exactly orthogonal if LISA is a perfect equilateral triangle and if all secondary noise sources, i.e., noises that dominate the LISA data stream after laser noise suppression through TDI, such as Test Mass (TM) acceleration and Optical Metrology System (OMS) noise, are perfectly equal in all three spacecraft. In fact, for realistic motion of the satellites, the arm-lengths will be unequal (and time-varying) at the percent level [24], and LISA will not be a perfectly equilateral triangle. Moreover, the noises appearing in the different spacecraft will not be equal, breaking another assumption underlying the idealized derivation of the A, E, and T variables.…”