Objective: To compare the differences of hip offset and rotation center reconstruction between robot-assisted and manual total hip arthroplasty (THA). Methods: Patients underwent robot-assisted and manual THA from May to September of 2020 in the First Affiliated Hospital of Chongqing Medical University were enrolled in this study. The patients included 27 patients (28 hips) in robot-assisted THA (rTHA) group and 29 patients (31 hips) in manual THA (mTHA) group. In rTHA group, there were 16 males and 11 females, with a mean age of (59±13) years. In mTHA group, there were 18 males and 11 females, with a mean age of (63±14) years. Basic information, including gender, age, body mass index (BMI), diagnosis and functional scoring etc, were recorded. In rTHA group, Mako robot system was used for preoperative planning, intraoperative real-time location and navigation. In mTHA group, traditional preoperative template design and surgical procedure were carried out. Operation time and functional scoring were compared postoperatively. Femoral offset, acetabular offset, global offset, rotation center changes in vertical and horizontal directions were measured on pelvis X-ray and analyzed. The correlation between intraoperative feedback of global offset change in robot system and postoperative measured global offset were analyzed. Results: Operation time in rTHA group was (80±10) min, which was statistically longer than that in mTHA group ((58±18) min, P<0.001). With 6 months' follow-up, the Harris scoring in rTHA group was 94.9±2.8, which was statistically higher than that in mTHA group (93.1±2.8, P=0.017), however there was no statistic difference in WOMAC scoring between rTHA and mTHA group (7.0±3.8 vs 7.1±2.4, P=0.840). Absolute global offset change within 5 mm, 5-10 mm and lager than 10 mm were 71.4%(20/28), 28.6%(8/28) and 0 in rTHA group, which were 45.2%(14/31), 29.0%(9/31) and 25.8%(8/31) in mTHA group (all P<0.05). A positive relation was found between intraoperative feedback of global offset change in robot system and postoperative measured global offset in rTHA group (r=0.77, P<0.001). It was found that rotation center changes concentrated in outer upper quadrant in both groups, and rotation center change in rTHA group concentrated mainly in the area less than 10 mm, however, rotation center change in mTHA group was more dispersive compared with rTHA group. Conclusion: rTHA may accurately reconstruct hip offset and rotation center, intraoperation feedback of global offset change may be an effective reference.