In　this　paper,　an　improved　trajectory　tracking　control　law　is　proposed　for　a　four-wheel　differential　nonholonomic　drive　intelligent　vehicle.　Based　on　the　kinematics　model　and　Euler-Lagrange　(EL)　system　of　the　vehicle,　the　control　laws　adopt　an　improved　typical　backstepping　method　to　design　the　kinematics　controller　and　a　fast　terminal　sliding　mode　technique　to　devise　the　dynamic　control　law,　which　can　completely　track　the　desired　trajectory　in　finite　time.　The　system　is　global　asymptotic　stability　by　Lyapunov　theory　and　the　parameters　are　optimized　through　the　pole　assignment　method　when　the　vehicle　is　near　the　reference　trajectory.　Finally,　simulations　demonstrate　the　effectiveness　and　correctness　of　the　proposed　control　strategy.