Background Several studies have shown that robotic devices can effectively improve motor function in stroke patients through limb activation. However, the effects of robot-assisted therapy on perceptual deficits after stroke is unclear. Objective This review aimed to evaluate the effectiveness of robotic limb activation in patients with unilateral spatial neglect (USN) after stroke. Methods In this systematic review, a literature search was performed using MEDLINE, EMBASE, CENTRAL, CINAHL, and LILACS databases without language restrictions. Randomized controlled trials (RCTs) and quasi-RCTs of robot-assisted therapy for USN after stroke were selected. Two reviewers independently assessed the risk of bias and certainty of the evidence of the included studies. Results A total of 630 studies were identified, including five studies for qualitative synthesis and four meta-analyses. The results of RCTs comparing robotic limb activation with a control group suggested an improvement in the degree of USN measured by the line bisection test (standardized mean difference [SMD], −0.64; 95% confidence interval [CI], −1.13 to −0.15; P = .01). There were no differences between the groups in the motor-free visual perception test 3rd edition (SMD, 0.27; 95% CI, −0.25–0.79; P = .31), star cancellation test (SMD, 0.26; 95% CI, −0.42−0.94; P = .54), Albert’s test (SMD, −0.67; 95% CI, −2.01−0.66; P = .32), and Catherine Bergego Scale (SMD, −0.81; 95% CI, −2.07−0.45; P = .21). Conclusion The study demonstrated that limb activation through robotic therapy can improve midline perception. However, there was no impact on tasks assessing visual scanning, functionality, or activities of daily living.