Robot-Assisted Gait Training (RAGT) has been shown to improve walking function in hemiparetic stroke. It is assumed, but unproven, that these improvements are associated with enhanced muscle coordination resulting from neurological changes in locomotor control. The goal of this study is to assess changes in muscle coordination in the lower extremities before and after an RAGT intervention using the Autoambulator TM . Four individuals with subacute stroke participated in this prospective case series. All participants had hemiparesis and were able to walk with supervision or minimal contact assistance. Each participant received 18 one-hour sessions of RAGT over an 8-week period. Before and after the RAGT intervention, gait was assessed using a Timed Up-and-Go and a 10 m Walk Test. Participants also underwent a muscle coordination evaluation based on surface electromyography (SEMG) recorded from lower extremity muscles. SEMG electrodes were placed to record the following major muscle groups bilaterally: vastus lateralis, biceps femoris, tibialis anterior and lateral gastrocnemius. A Symmetry Index (SI) was defined to represent the similarity of SEMG signals from both sides of the body during gait. As well, an Index of Rhythmicity (IR) was defined to represent the degree to which the composite SEMG signals could be characterized by a basic recurring pattern. Following the RAGT intervention, participants demonstrated a 16 to 60% increase in walking speed and a 14 to 37% decrease in time to complete the Timed Up-and-Go test. Similarly, all four participants showed improvements in SI and IR. These results indicate that in addition to significant improvements in walking function following RAGT, muscle activation patterns were more rhythmic and more coordinated between both sides of the body. These findings suggest that the improvements in gait function following RAGT are associated with improvements in muscle coordination. These changes are likely due to positive adaptations in the central nervous system.