The　conventional　A　star　algorithm　is　one　of　the　most　effective　direct　search　methods　to　solve　the　shortest　path　in　the　static　map.　However,　to　keep　the　path　as　short　as　possible,　the　best　path　it　seeks　tends　to　be　close　to　obstacles,　which　has　defects　for　some　medical　and　service　robots　that　need　to　keep　a　safe　distance　from　obstacles.　In　this　paper,　based　on　the　conventional　A　star　algorithm,　the　computational　form　of　its　heuristic　function　is　adjusted　so　that　the　search　efficiency　is　improved　by　nearly　80%.　Meanwhile,　the　obstacles　are　extended　during　calculation　so　that　they　are　kept　in　line　with　the　obstacles.　Experiments　show　that　the　mentioned　method　can　improve　computational　efficiency　while　ensuring　that　the　generated　path　and　obstacles　are　within　the　safe　range.　The　results　of　this　paper　may　be　conducive　to　the　application　of　the　algorithm　in　practical　situations.　©　2023　IEEE.