We conducted a systematic review with meta-analysis to determine the evidence in support of exercise to improve sleep quality assessed subjectively and objectively in Parkinson's Disease (PD). Standardized mean differences (SMD) comparing the effects of exercise and control interventions on sleep quality with 95% confidence intervals (CI) were calculated. Data from 10 randomized and 2 non-randomized controlled trials, including a total of 690 persons with PD were included. Exercise had a significant positive effect on sleep quality assessed subjectively (SMD = 0.53; 95% CI = 0.16–0.90; p = 0.005). However, the methodological quality of the studies showing positive effects on sleep quality was significantly poorer than the studies showing no effects. Only one study assessed the impact of exercise on objective sleep quality, showing improvements in sleep efficiency assessed with polysomnography (SMD = 0.94; 95% CI = 0.38–1.50; p = 0.001). Exercise performed at moderate to maximal intensities (SMD = 0.46; 95% CI = 0.05–0.87; p = 0.03) had significant effects on subjective sleep quality. In contrast, exercise performed at mild to moderate intensities showed non-significant effects (SMD = 0.76; 95% CI = −0.24–1.76; p = 0.14). These results support the use of exercise to improve sleep quality in persons with PD and reinforce the importance of achieving vigorous exercise intensities. Biases, limitations, practice points and directions for future research are discussed.