To　ensure　the　high　quality　and　stability　of　oral　diagnosis　and　treatment　images.　It　is　required　that　the　position　of　the　X-ray　ball　tube　is　accurate　during　the　shooting　process.　The　artificial　difference　should　be　eliminated.　In　order　to　avoid　cross　infection　between　doctors　and　patients,　an　overall　solution　of　intra-oral　radiograph　assisted　robot　(IRAR)　was　proposed,　including　face　recognition,　camera　calibration,　end　effector　spatial　positioning　and　robot-oral　interaction.　The　working　environment　and　mechanism　kinematics　of　IRAR　were　analyzed.　Based　on　the　establishment　of　binocular　vision　calibration　system,　a　face　recognition　algorithm　aided　image　method　was　proposed.　The　interaction　model　between　virtual　image　and　robot　pose　was　analyzed.　By　analyzing　the　function　of　the　robot　end　effector,　the　related　parameter　requirements　of　the　end　effector　were　established.　Finally,　the　imaging　work　area　planning　scheme　of　the　imaging　robot　was　obtained.　Through　the　interaction　between　the　robot　and　the　oral　cavity,　the　non-contact　dental　medical　images　were　photographed　and　disinfected.　The　repeated　positioning　accuracy　of　IRAR　system　shall　not　be　greater　than　1　mm.　The　average　error　is　0.5 mm.　The　experiment　proved　that　IRAR　can　finish　the　shooting　task　smoothly　and　accurately.　©　2022,　The　Author(s),　under　exclusive　license　to　Springer　Nature　Switzerland　AG.