To evaluate whether robot-assisted gait training combined with transcranial direct current stimulation is more effective than robot-assisted gait training alone or conventional walking rehabilitation for improving walking ability in stroke patients. Pilot randomized clinical trial. Rehabilitation unit of a university hospital. Thirty patients with chronic stroke. All patients received ten 50-minute treatment sessions, five days a week, for two consecutive weeks. Group 1 (n = 10) underwent a robot-assisted gait training combined with transcranial direct current stimulation; group 2 (n = 10) underwent a robot-assisted gait training combined with sham transcranial direct current stimulation; group 3 (n = 10) performed overground walking exercises. Patients were evaluated before, immediately after and two weeks post treatment. six-minute walking test, 10-m walking test. No differences were found between groups 1 and 2 for all primary outcome measures at the after treatment and follow-up evaluations. A statistically significant improvement was found after treatment in performance on the six-minute walking test and the 10-m walking test in favour of group 1 (six-minute walking test: 205.20 ± 61.16 m; 10-m walking test: 16.20 ± 7.65 s) and group 2 (six-minute walking test: 182.5 ± 69.30 m; 10-m walking test: 17.71 ± 8.20 s) compared with group 3 (six-minute walking test: 116.30 ± 75.40 m; 10-m walking test: 26.30 ± 14.10 s). All improvements were maintained at the follow-up evaluation. In the present pilot study transcranial direct current stimulation had no additional effect on robot-assisted gait training in patients with chronic stroke. Larger studies are required to confirm these preliminary findings.