Expected performance of the ATLAS detector under different High-Luminosity LHC conditions

A major upgrade of the LHC is planned to be completed by 2027, resulting in the so-called High-Luminosity LHC (HL-LHC) operating at a target luminosity of $5\times10^{34}~\mathrm{cm}^{-2}\mathrm{s}^{-1}$, with a possibility to reach an ultimate peak luminosity of up to $7.5\times10^{34}~\mathrm{cm}^{-2}\mathrm{s}^{-1}$. This corresponds to an average of 140 to 200 proton-proton collisions per bunch crossing. To help the HL-LHC experiments separate particles coming from the different collisions in a single crossing, the HL-LHC can manipulate the beams in various ways to change the density of these collisions in the spatial dimensions, at the expense of potential performance degradation and financial costs. This note documents studies of the performance of the object-level reconstruction and identification with the ATLAS detector at HL-LHC as a function of the number of collisions per bunch crossing and their density.