Expected performance of the upgrade ATLAS experiment for HL-LHC

The Large Hadron Collider (LHC) has been successfully delivering proton-proton collision data at the unprecedented center of mass energy of 13 TeV. An upgrade is planned to increase the instantaneous luminosity delivered by the LHC in what is called the HL-LHC, aiming to deliver a total of up 3000 fb$^{-1}$ to 4000 fb$^{-1}$ of data per experiment. To cope with the expected data-taking conditions ATLAS is planning upgrades of the detector. Six Technical Design Reports (TDR) were produced by the ATLAS Collaboration. In these TDRs the physics motivation and benefits of such upgrades are discussed together with details on the upgrade project itself. In this contribution we review the expected performance of the upgraded ATLAS detector and the expected reach for physics measurements as well as the discovery potential for new physics that is expected by the end of the HL-LHC data-taking. The performance of object reconstruction under the expected pile-up conditions will be shown, including a fully re-optimized $b$-tagging algorithm. Important benchmark physics projections including di-Higgs boson production sensitivity will be discussed