Due　to　the　expiration　of　the　design　service　life　of　some　nuclear　facilities,　a　large　number　of　hazardous　radioactive　devices　need　to　be　dismantled.　Therefore,　special　attention　needs　to　be　paid　to　the　safety　and　reliability　of　the　dismantling　process　to　ensure　the　protection　of　workers　and　the　environment.　In　response　to　this　challenge,　this　article　proposes　an　innovative　solution:　a　9-DOF　(Degrees　of　Freedom)　nuclear　decommissioning　robot　composed　of　a　3-DOF　gantry　and　a　6-DOF　robotic　arm.　The　core　of　this　robot　design　lies　in　utilizing　its　high　redundancy　of　9　DOFs　and　the　intelligence　of　machine　learning　to　perform　complex　dismantling　tasks.　At　the　same　time,　to　ensure　the　accuracy　and　safety　of　the　operation,　a　forward　and　inverse　kinematics　analysis　of　the　robot　is　required　to　gain　a　detailed　understanding　of　its　motion　capabilities　and　limitations,　making　it　more　suitable　for　dismantling　work　in　nuclear　facilities.　The　ultimate　goal　of　this　research　is　to　effectively　recycle　graphite　reactors.　Experiments　show　that　the　method　of　using　this　9-DOF　nuclear　decommissioning　robot　combined　with　the　RRT　(Rapidly-exploring　Random　Tree)　algorithm　for　path　planning　has　certain　feasibility　in　practical　applications.　This　not　only　demonstrates　the　theoretical　effectiveness　of　this　technical　solution,　but　also　provides　important　reference　value　for　future　practical　applications.　Through　this　method,　we　can　safely　and　effectively　handle　decommissioned　nuclear　facilities,　thus　reducing　the　potential　threat　of　nuclear　radiation　to　the　environment　and　human　health.　©　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.　2025.