The Kepler mission has shown that small planets are extremely common. It is likely that nearly every star in the sky hosts at least one rocky planet. We just need to look hard enough - but this requires vast amounts of telescope time. MINERVA (MINiature Exoplanet Radial Velocity Array) is a dedicated exoplanet observatory with the primary goal of discovering rocky, Earth-like planets orbiting in the habitable zone of bright, nearby stars. The MINERVA team is a collaboration among UNSW Australia, Harvard-Smithsonian Center for Astrophysics, Penn State University, University of Montana, and the California Institute of Technology. The four-telescope MINERVA array will be sited at the F.L. Whipple Observatory on Mt Hopkins in Arizona, USA. Full science operations will begin in mid-2015 with all four telescopes and a stabilised spectrograph capable of high-precision Doppler velocity measurements. We will observe ~100 of the nearest, brightest, Sun-like stars every night for at least five years. Detailed simulations of the target list and survey strategy lead us to expect 154 new low-mass planets.

We have investigated the dynamical stability of the proposed companions orbiting the Algol type short-period eclipsing binary SW Lyncis (Kim et al. 2010). The two candidate companions are of stellar to substellar nature, and were inferred from timing measurements of the system’s primary and secondary eclipses. We applied well-tested numerical techniques to accurately integrate the orbits of the two companions and to test for chaotic dynamical behavior. We carried out the stability analysis within a systematic parameter survey varying both the geometries and orientation of the orbits of the companions, as well as their masses. In all our numerical integrations we found that the proposed SW Lyn multi-body system is highly unstable on time-scales on the order of 1000 years. Our results cast doubt on the interpretation that the timing variations are caused by two companions. This work demonstrates that a straightforward dynamical analysis can help to test whether a best-fit companion-based model is a physically viable explanation for measured eclipse timing variations. We conclude that dynamical considerations reveal that the proposed SW Lyncis multi-body system most likely does not exist or the companions have significantly different orbital properties from those conjectured in Kim et al. (2010).