Using　Unmanned　Aerial　Vehicles　as　mobile　base　stations　is　a　promising　way　to　collect　data　from　sensor　nodes,　especially　for　large-scale　wireless　sensor　networks.　Previous　works　mainly　focus　on　improving　the　freshness　of　the　collected　data　or　the　energy　efficiency　by　scheduling　UAVs.　Considering　the　fact　that　the　sensing　data　in　some　applications　is　time-sensitive,　that　is,　the　value　of　the　sensing　data　is　based　on　its　Timeliness　of　Information　(ToI),　which　decays　over　time.　Therefore,　in　this　paper,　we　investigate　the　UAV　Trajectory　optimization　problem　for　Maximizing　the　ToI-based　data　utility　(TMT).　We　propose　an　improved　deep　reinforcement　learning-based　algorithm　to　address　the　problem,　and　the　experience　results　demonstrate　the　effectiveness　of　our　designs.　©　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.　2024.