To compare the safety and nail placement accuracy of fluoroscopy-assisted and robot-assisted minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) in the treatment of single-space lumbar disc herniation. The clinical data of 52 patients with single-space lumbar disc herniation treated by MIS-TLIF from March 2019 to February 2020 were retrospectively analyzed. Among them, 24 patients were treated by robot-assisted MIS-TLIF(group A) and 28 patients were treated by fluoroscopy-assisted MIS-TLIF (group B). The intraoperative blood loss, operation time, intraoperative fluoroscopy times, preoperative and postoperative visual analogue scale(VAS), Japanese Orthopaedic Association(JOA) scores and operation-related complications were recorded in two groups. Gertzbein-Robbins grade according to CT scan was used to evaluate the nail placement after operation. Grade A and B were evaluated as satisfactory nail placement, and grade C, D, and E were evaluated as error placement. Babu's method was used to evaluate the screw's invasion to the superior articular process. The operation time, intraoperative blood loss and intraoperative fluoroscopy times in group A were less than those in group B(P<0.05).VAS and JOA scores of all patients at the final follow-up were significantly improved compared with those before operation(P<0.05), but there was no statistically significant difference between the groups(P>0.05). There were 96 and 112 screws in group A and group B, respectively. Three days after operation, according to the Gertzbein-Robbins grade to evaluate the nail placement accuracy, there were 90 screws of grade A, 5 of grade B, 1 of grade C, no grade D and E in group A;there were 84 screws of grade A, 16 of grade B, 8 of grade C, 4 of grade D, no grade E in group B;the difference between two groups was statistically significant(Z=-3.709, P=0.000). The satisfactory rate of screw placement in group A was 98.96% (95/96), and that of group B was 89.29% (100/112), the difference between two groups was statistically significant (χ2=8.254, P=0.004). Three days after operation, the invasion of superior facet joints by pedicle screws was evaluated according to Babu's method, including 90 screws in grade 0, 4 in grade 1, 2 in grade 2, and 0 in grade 3 in group A;86 in grade 0, 12 in grade 1, 10 in grade 2 and 4 in grade 3 in group B, and the difference was statistically significant(Z=-3.433, P=0.001). There were no serious spinal cord, nerve and vascular injuries and other operation-related complications caused by screw implantation failure in both groups. All patients were followed up from 6 to 12(9.06±1.60) months. The neurological symptoms improved well after operation. During the follow-up period, there was no recurrence of symptoms, loosening or breakage of the internal fixation. Compared with the traditional fluoroscopy-assisted MIS-TLIF, the spinal robot-assisted MIS-TLIF not only has more minimally invasive and safer, but also has higher accuracy in nail placement, lower incidence of upper articular process invasion, and more accurate decompression targets, which can be used for minimally invasive treatment of single-space lumbar disc herniation.