Objective To investigate the feasibility and effectiveness of robot-assisted posterior minimally invasive access in treatment of thoracolumbar tuberculosis via transforaminal expansion approach. Methods A clinical data of 40 patients with thoracolumbar tuberculosis admitted between January 2017 and May 2022 and met the selection criteria was retrospectively analyzed. Among them, 15 cases were treated with robot-assisted and minimally invasive access via transforaminal expansion approach for lesion removal, bone graft, and internal fixation (robotic group), and 25 cases were treated with traditional transforaminal posterior approach for lesion removal and intervertebral bone grafting (traditional group). There was no significant difference in the baseline data between the two groups (P>0.05) in terms of gender, age, lesion segment, and preoperative American Spinal Injury Association (ASIA) grading, Cobb angle, visual analogue scale (VAS) score, erythrocyte sedimentation rate (ESR), and C reactive protein (CRP). The outcome indicators were recorded and compared between the two groups, including operation time, intraoperative bleeding volume, hospital stay, postoperative bedtime, complications, ESR and CRP before operation and at 1 week after operation, the level of serum albumin at 3 days after operation, VAS score and ASIA grading of neurological function before operation and at 6 months after operation, the implant fusion, fusion time, Cobb angle of the lesion, and the loss of Cobb angle observed by X-ray films and CT. The differences of ESR, CRP, and VAS score (change values) between pre- and post-operation were calculated and compared. Results Compared with the traditional group, the operation time and intraoperative bleeding volume in the robotic group were significantly lower and the serum albumin level at 3 days after operation was significantly higher (P<0.05); the postoperative bedtime and the length of hospital stay were also shorter, but the difference was not significant (P>0.05). There were 2 cases of poor incision healing in the traditional group, but no complication occurred in the robotic group, and the difference in the incidence of complication between the two groups was not significant (P>0.05). There were significant differences in the change values of ESR and CRP between the two groups (P<0.05). All Patients were followed up, and the follow-up time was 12-18 months (mean, 13.0 months) in the traditional group and 12-16 months (mean, 13.0 months) in the robotic group. Imaging review showed that all bone grafts fused, and the difference in fusion time between the two groups was not significant (P>0.05). The difference in Cobb angle between the pre- and post-operation in the two groups was significant (P<0.05); and the Cobb angle loss was significant more in the traditional group than in the robotic group (P<0.05). The VAS scores of the two groups significantly decreased at 6 months after operation when compared with those before operation (P<0.05); the difference in the change values of VAS scores between the two groups was not significant (P>0.05). There was no occurrence or aggravation of spinal cord neurological impairment in the two groups after operation. There was a significant difference in ASIA grading between the two groups at 6 months after operation compared to that before operation (P<0.05), while there was no significant difference between the two groups (P>0.05). Conclusion Compared with traditional posterior open operation, the use of robot-assisted minimally invasive access via transforaminal approach for lesion removal and bone grafting internal fixation in the treatment of thoracolumbar tuberculosis can reduce the operation time and intraoperative bleeding, minimizes surgical trauma, and obtain definite effectiveness.