Motor impairment is the most common symptom in multiple sclerosis (MS). Thus, a variety of new rehabilitative strategies, including robotic gait training, have been implemented, showing their effectiveness. The aim of our study was to investigate whether an intensive robotic gait training, preceding a traditional rehabilitative treatment, could be useful in improving and potentiating motor performance in MS patients. Forty-five patients, who fulfilled the inclusion criteria, were enrolled in this study and randomized into either the control group (CG) or the experimental group (EG). A complete clinical evaluation, including the Expanded Disability Severity Scale, the Functional Independence Measure, the Hamilton Rating Scale for Depression, the time up and go test (TUG), and the Tinetti balance scale, was performed at baseline (T0), after 6 week (T1), at the end of rehabilitative training (T2), and 1 month later (T3). A significant improvement was observed in the EG for all the outcome measures, whereas the CG showed an improvement only in TUG. In contrast, from T1 to T2, only CG significantly improved in all outcomes, whereas the EG had an improvement only regarding TUG. From T2 to T3, no significant differences in Functional Independence Measure scores emerged for both the groups, but a significant worsening in Tinetti balance scale and TUG was observed for the CG and in TUG for the EG. Our study provides evidence that robotic rehabilitationn coupled with two-dimensional virtual reality may be a valuable tool in promoting functional recovery in patients with MS.