To systematically evaluate the effects of robot-assisted rehabilitation on motor and balance functions, and basic activities of daily living in patients with cerebral palsy. Methods: A comprehensive computerized search of Chinese and foreign language databases was performed to identify randomized controlled trials (RCTs) of robot-assisted rehabilitation for cerebral palsy. Data were extracted from clinical studies that met the inclusion criteria. Quality evaluation was performed using the Cochrane Handbook for Systematic Reviews of Interventions version 5.0, and meta-analysis was performed using RevMan 5.4 statistical software. Results: A total of 12 RCTs with 797 patients were included. The meta-analysis results showed that, compared with traditional rehabilitation methods, robot-assisted rehabilitation significantly improved the balance function score [standard mean difference (SMD)=0.67, 95% CI 0.16～1.18), P=0.010], the gross motor function measure-D area score (GMFM-D) [SMD=0.70, 95% CI 0.28～1.12, P=0.001], gross motor function measure-E area score (GMFM-E) [SMD=0.50, 95% CI 0.26～0.75, P<0.0001], and the basic activities of daily living [MD=8.77, 95% CI 2.64～14.89, P=0.005] in patients with cerebral palsy. Conclusion: Compared with traditional rehabilitation methods, robot-assisted rehabilitation can significantly improve motor and balance function scores, and basic activities of daily living in patients with cerebral palsy.