Objective: To compare the clinical and radiographic outcomes between robot-assisted minimally invasive transforaminal lumbar interbody fusion (RA MIS-TLIF) and fluoroscopy-assisted MIS-TLIF (FA MIS-TLIF) in patients with degenerative lumbar spinal diseases. Methods: The clinical data of 114 patients with lumbar degenerative diseases who underwent MIS-TLIF in Qilu Hospital of Shandong University from January 2019 to March 2022 were analyzed retrospectively. Fifty-eight patients underwent RA MIS-TLIF (robot group) and 56 patients received FA MIS-TLIF (freehand group). There were 16 males and 42 females in the robot group, with a mean age of (56.7±8.1) years. And there were 19 males and 37 females in the freehand group, with a mean age of (57.2±8.6) years. The clinical outcome parameters were the visual analog scale (VAS) of pain, Oswestry Disability Index (ODI) score, operative time, number of intraoperative fluoroscopies, blood loss, postoperative hospital stay and complications. The radiographic change measures were the accuracy of screw placement, facet joint violation (FJV), fusion status, and change in disc height at the proximal adjacent segment. All the patients were followed-up for 2-5 years. Results: There was no significant differences in the VAS and ODI scores, blood loss, or postoperative hospital stay between the two groups (all P>0.05). The operative time was longer in robot group than freehand group [(158.5±12.1) min vs (146.4±15.4) min, P<0.001]. There was no significant difference in the number of intraoperative fluoroscopies for patients between robot group and freehand group (P>0.05). The number of intraoperative fluoroscopies for the surgeon was significantly lower in robot group than freehand group (13.8±3.9 vs 74.7±6.8, P<0.001). The rate of a perfect screw position (grade A) was higher in robot group than freehand group [87.5%(203/232) vs 70.1%(157/224), P<0.001]. However, there was no significant difference in the proportion of clinically acceptable screws (grades A and B) between the two groups [98.3%(228/232) vs 96.9%(217/224), P=0.330]. The FJV grade was significantly higher in freehand group than robot group (0.43±0.68 vs 0.13±0.43, P<0.001). During at 2-year postoperative follow-up, there was no significant difference in the fusion status between the two groups (P>0.05); however, the decrease in disc height at the proximal adjacent segment was significantly less in robot group than freehand group [(0.63±0.38) mm vs (0.92±0.35) mm, P=0.001]. In the robotic group, a pedicle screw penetrated the outer wall of the vertebral pedicle in one patient, which was adjusted during surgery. In the freehand group, two screws were inserted too deeply and penetrated the anterior cortex, resulting in mild abdominal discomfort postoperatively, which resolved by the third day after surgery. Conclusions: Robot-assisted percutaneous pedicle screw placement is a safer and more accurate alternative to conventional freehand fluoroscopy-assisted pedicle screw insertion in MIS-TLIF. Compared with freehand MIS-TLIF, robot-assisted MIS-TLIF increases the operation time, but the accuracy of screw placement is higher, and the intraoperative radiation dose and the degree of adjacent segment degeneration are reduced.