To　improve　the　energy-regenerative　efficiency　and　robustness　of　electric　vehicle　(EV),　a　novel　energy-regenerative　controller　was　designed　and　applied　to　the　charge　current　loop　of　the　EV.　The　controller　that　combines　neural　network　(NN)　with　traditional　sliding　mode　controller　(SMC)　comprises　a　radial　basis　function　NN　(RBFNN)　and　a　SMC.　The　RBFNN　is　used　to　adaptively　adjust　the　switching　gain　of　the　SMC.　The　simulation　model　of　the　energy-regenerative　system　is　built　with　MATLAB/SIMULINK,　and　the　simulation　results　show　that　comparing　with　traditional　SMC,　the　NNSMC　has　better　performance　at　response　speed,　steady-state　tracking　error　and　resisting　disturbance　in　energy-regenerative　process.　Additionally,　it　can　recover　more　energy.