Based　on　the　intrinsic　motivation　mechanism,　this　paper　builds　up　a　new　kind　of　path　planning　method.　The　curiosity　function　existing　in　such　a　mechanism　is　extended　from　state-based　to　motion-based,　and　the　corresponding　motion　choosing　mechanism　is　designed,　which　is　more　biologically　explicable.　The　energy　function　of　the　mobile　robot　under　the　environment　with　obstacles　is　designed,　which　is　used　to　decide　the　learning　direction.　Simulation　results　show　that　the　proposed　path　planning　method　can　help　the　robot　learn　its　environment,　and　achieve　obstacle-free　navigation　from　any　start　position.　What　is　more,　the　energy　function　is　environment　independent,　and　even　though　the　target　position　changes,　the　robot　can　still　complete　the　tasks　through　relearning,　which　proves　that　the　proposed　method　is　with　high　autonomy　and　not　task　specific.　©　2018,　Editorial　Office　of　Control　and　Decision.　All　right　reserved.