Terrestrial planet formation theory is at a bottleneck, with the growing realization that pairwise collisions are treated far too simply. Here, and in our companion paper (Cambioni et al. 2019) that introduces the training methodology, we demonstrate the first application of machine learning to more realistically model the late stage of planet formation by giant impacts. We present surrogate models that give fast, reliable answers for the masses and velocities of the two largest remnants of a giant impact, as a function of the colliding masses and their impact velocity and angle, with the caveat that our training data do not yet include pre-impact rotation or variable thermal conditions. We compare canonical N-body scenarios of terrestrial planet formation assuming perfect merger (Chambers 2001) with our more realistic treatment that includes inefficient accretions and hit-and-run collisions. The result is a protracted tail of final events lasting ~200 Myr, and the conversion of about half the mass of the initial population to debris. We obtain profoundly different solar system architectures, featuring a much wider range of terrestrial planet masses and enhanced compositional diversity.

Planet formation simulations are capable of directly integrating the evolution of hundreds to thousands of planetary embryos and planetesimals, as they accrete pairwise to become planets. In principle such investigations allow us to better understand the final configuration and geochemistry of the terrestrial planets, as well as to place our solar system in the context of other exosolar systems. These simulations, however, classically prescribe collisions to result in perfect mergers, but computational advances have begun to allow for more complex outcomes to be implemented. Here we apply machine learning to a large but sparse database of giant impact studies, streamlining simulations into a classifier of collision outcomes and a regressor of accretion efficiency. The classifier maps a 4-Dimensional parameter space (target mass, projectile-to-target mass ratio, impact velocity, impact angle) into the four major collision types: merger, "graze-and-merge", "hit-and-run", and disruption. The definition of the four regimes and their boundary is fully data-driven; the results do not suffer from any model assumption in the fitting. The classifier maps the structure of the parameter space and provides insights about the outcome regimes. The regressor is a neural network which is trained to closely mimic the functional relationship between the 4-D space of collision parameters, and a real-variable outcome, the mass of the largest remnant. This work is a prototype of a more complete surrogate model, based on extended sets of simulations ("big data"), that will quickly and reliably predict specific collision outcomes for use in realistic N-body dynamical studies of planetary formation.

We present a new method for the determination of the surface properties of airless bodies from measurements of the emitted infrared flux. Our approach uses machine learning techniques to train, validate, and test a neural network representation of the thermophysical behavior of the atmosphereless body given shape model, illumination and observational geometry of the remote sensors. The networks are trained on a dataset of thermal simulations of the emitted infrared flux for different values of surface rock abundance, roughness, and values of the thermal inertia of the regolith and of the rock components. These surrogate models are then employed to retrieve the surface thermal properties by Markov Chain Monte Carlo Bayesian inversion of observed infrared fluxes. We apply the method to the inversion of simulated infrared fluxes of asteroid (101195) Bennu -- according to a geometry of observations similar to those planned for NASA's OSIRIS-REx mission -- and infrared observations of asteroid (25143) Itokawa. In both cases, the surface properties of the asteroid -- such as surface roughness, thermal inertia of the regolith and rock component, and relative rock abundance -- are retrieved; the contribution from the regolith and rock components are well separated. For the case of Itokawa, we retrieve a rock abundance of about 85% for pebbles larger than the diurnal skin depth, which is about 2 cm. The thermal inertia of the rock is found to be lower than the expected value for LL chondrites, indicating that the rocks on Itokawa could be fractured. The average thermal inertia of the surface is around 750 $J s^{-1/2} K^{-1} m^{-2}$ and the measurement of thermal inertia of the regolith corresponds to an average regolith particle diameter of about 10 mm, consistently with in situ measurements as well as results from previous studies.

We measure the mid-plane of the main asteroid belt by using the observational data of a nearly complete and unbiased sample of asteroids, and find that it has inclination $\bar{I}=0.93\pm0.04$ degrees and longitude of ascending node $\bar{\Omega}=87.6\pm2.6$ degrees (in J2000 ecliptic-equinox coordinate system). This plane differs significantly from previously published measurements, and it is also distinctly different than the solar system's invariable plane as well as Jupiter's orbit plane. The mid-plane of the asteroid belt is theoretically expected to be a slightly warped sheet whose local normal is controlled by the gravity of the major planets. Specifically, its inclination and longitude of ascending node varies with semi-major axis and time (on secular timescales), and is defined by the forced solution of secular perturbation theory; the $\nu_{16}$ nodal secular resonance is predicted to cause a significant warp of the mid-plane in the inner asteroid belt. We test the secular theory by measuring the current location of the asteroids' mid-plane in finer semi-major axis bins. We find that the measured mid-plane in the middle and outer asteroid belt is consistent, within 3--$\sigma$ confidence level, with the prediction of secular perturbation theory, but a notable discrepancy is present in the inner asteroid belt near $\sim2$ AU.