Under　the　condition　that　the　base　position　and　attitude　of　the　free-floating　dual-arm　space　robot　system　are　uncontrolled,　the　self-learning　control　problem　based　on　fuzzy　Gaussian　function　neural　network　is　addressed.　Since　this　space　robot　strictly　abide　by　the　rules　of　momentum　conservation　and　angular　momentum　conservation,　its　dynamic　equations　exhibit　a　strong　nonlinearity.　When　fuzzy　control　and　neural　network　are　combined,　i.e.,　the　fuzzy　inference　is　realized　by　neural　network,　the　fuzzy　control　obtains　its　self-learning　ability.　Then　the　self-teaming　control　scheme　based　on　fuzzy　Gaussian　function　neural　network　is　designed　for　the　dual-arm　space　robot　system　in　joint　space.　Numerical　simulation　is　carried　out,　and　the　effectiveness　of　the　proposed　method　is　validated.