The　task　of　exploring　unknown　environments　using　autonomous　mobile　robots　is　essential.　This　paper　is　based　on　the　internal　motivation　mechanism　combined　with　the　Shannon　entropy　theory.　Considering　that　the　current　map　may　generate　important　information　gains　about　the　environment,　a　navigation　strategy　for　effectively　detecting　unknown　environments　is　proposed.　The　strategy　is　based　on　the　local　information　in　the　currently　established　map,　so　that　the　robot　could　predict　the　unknown　environment,　uses　the　Shannon　entropy　to　estimate　the　information　gain　of　each　frontier　to　be　selected,　and　uses　the　information　gain　as　the　internal　motivation　of　the　robot　to　make　the　robot　choose　a　frontier　with　the　largest　information　gain　to　reach.　The　experimental　results　show　that　the　proposed　method　can　effectively　help　the　robot　to　explore　the　unknown　environment,　realize　environment　cognition,　and　establish　a　complete　and　accurate　environmental　map.　©　2020　IEEE.