In　this　paper,　a　nonsingular　terminal　sliding　mode　control　(NTSMC)　combined　with　neural　network　and　disturbance　observer　is　proposed　for　the　trajectory　tracking　of　an　underactuated　AUV.　An　adaptive　neural　network　(NN)　is　designed　to　compensate　the　complicated　uncertain　nonlinearity　in　the　dynamics　of　AUV.　Disturbance　observer　is　employed　to　estimate　the　external　disturbances.　Lyapunov　method　proves　the　stability　of　the　proposed　controller.　Simulation　results　show　that　the　designed　controller　has　good　performance.　©　2023　IEEE.