The purpose of this study was to investigate the effects of robot-assisted gait training (RAGT) on mobility, activities of daily living (ADLs), and quality of life (QoL) in stroke rehabilitation. Fifty-one stroke patients randomly assigned to Group 1, Group 2, and Group 3 received conventional training (CT) plus RAGT, CT, and RAGT, respectively. The training duration was for 6 weeks. The primary outcome measures were the Barthel Index (BI), Stroke Specific Quality of Life Scale (SS-QOL), 6-Minute Walk Test (6-MWT), and Stair Climbing Test (SCT). The secondary outcomes were Fugl Meyer Assessment-Lower Extremity (FMA-LE), Comfortable 10-m Walk Test (CWT), Fast 10-m Walk Test (FWT), and Rate of Perceived Exertion (RPE). The mean change in all the primary [BI (p = 0.001), 6-MWT (p = 0.001), SS-QOL (p < 0.0001), and SCT (p = 0.004)] and except the FWT (p = 0.354) all the other secondary outcomes [FMA-LE (p = 0.049), CWT (p = 0.025) and RPE (p = 0.023)] improved significantly between the three groups. In the subgroup analysis, BI, 6-MWT, SS-QOL, and SCT improved significantly in Group 1 compared to Group 2 and Group 3 (p < 0.016). However, FMA-LE, CWT, and the RPE significantly improved in Group 1 compared to Group 2 and, also, only CWT improved significantly in Group 1 compared to Group 3 (p = 0.011). In a subgroup analysis of the primary and secondary outcome measures, there were no significant differences in Group 2 compared to Group 3 (p > 0.05). While combined training leads to more improvement in mobility, ADLs, and QoL, CT showed a similar improvement compared to the RAGT in stroke patients.