Study Objectives: We aimed (1) to investigate the effect of a 12-week exercise training on sleep quality and heart rate variability in middle-aged and older adults with poor sleep quality and (2) to examine the factors associated with the improvements in sleep quality and heart rate variability parameters. Methods: Forty adults aged ≥ 40 years with poor sleep quality (mean age = 62 years; 82.5% female) participated in this study. They were randomized into the exercise group or control group. Each exercise training program consisted of 40 minutes of supervised aerobic exercise training and 10 minutes of stretching class, 3 times a week for 12 weeks. Outcome measures included both subjective (Pittsburgh Sleep Quality Index) and objective (actigraphy recordings) sleep quality assessments, a cardiopulmonary exercise test, and heart rate variability assessment. Results: The exercise group showed significant improvements in the global score (P =.003), on all subscales of Pittsburgh Sleep Quality Index (P <.05), and in some heart rate variability parameters compared to the control group. Multiple regression analysis indicated that exercise participation was associated with either the sleep quality (β = −0.617, R 2 = .407; F = 6.226, P <.001) or heart rate monitor high frequency normalized units (β = 0.503, R 2 = .225; F = 3.200, P =.003) after adjustment for basic characteristics. However, the statistical significance between exercise participation and heart rate monitor high frequency normalized units diminished after controlling for the Pittsburgh Sleep Quality Index. Conclusions: Our results indicated that moderate-intensity exercise training had a beneficial effect on sleep quality and cardiac autonomic function. Middle-aged and older adults with poor sleep quality should be encouraged to engage in a moderate-intensity aerobic exercise training to improve their sleep quality and cardiac autonomic function.