Geometric　passability　parameters　are　key　for　the　terrain　adaptability　of　vehicles.　In　order　to　replace　the　current　empirically　manual　measurement　methods,　we　present　an　automatic　acquisition　method　for　measuring　vehicle　geometry　passability　parameters　with　high　quality　data.　A　remote　control　center　and　a　wheeled　robot　with　lidar　are　used　in　this　method　to　obtain　a　point　cloud　for　geometric　passability　parameters　calcutlation　at　the　bottom　of　vehicles.　The　data　acquisition　in　this　scene　is　more　likely　to　be　misregistered　by　the　instructor　due　to　the　majority　of　ground　points.　To　have　a　high　quality　of　point　cloud,　the　point　cloud　registration　algorithm　is　modified　by　appropriate　matching　point　distribution　and　pose　constraints.　The　experimental　results　show　that　this　method　performs　well　and　can　reliably　complete　the　data　acquisition　task　efficiently.　©　2021　IEEE