Background: Sensor-based technological therapy devices may be good candidates for neuromotor rehabilitation of people with Multiple Sclerosis (MS), especially for treating upper extremities function limitations. The sensor-based device rehabilitation is characterized by interactive therapy games with audio-visual feedback that allows training the movement of shoulders, elbows, and wrist, measuring the strength and the active range of motion of upper limb, registering data in an electronic database to quantitatively monitoring measures and therapy progress. Objective: This study aimed to investigate the effects of sensor-based motor rehabilitation in add-on to the conventional neurorehabilitation, on increasing the upper limb functions of patients with MS. Methods: Thirty patients were enrolled in the study and randomly assigned to the experimental group and the control group. The training consisting of twelve sessions of upper limb training was compared with twelve sessions of upper limb sensory-motor training, without robotic support. Both rehabilitation programs were performed for 40 minutes three times a week, for 4 weeks, in addition to conventional therapy. All patients were evaluated at the baseline (T0) and after 4 weeks of training (T1). Results: The within-subject analysis showed a statistically significant improvement in both groups, in the Modified Barthel Index and in the Rivermead Mobility Index scores and a significant improvement in Multiple Sclerosis Quality of Life-54 in the experimental. The analysis of effectiveness revealed that, compared with baseline (T0), the improvement percentage in all clinical scale scores was greater in the experimental group than the control group. Conclusions: Proposed training provides an intensive and functional-oriented rehabilitation that objectively evaluates achieved progress through exercises. Therefore, it can represent a good complementary strategy for hand rehabilitation in MS patients.