Objective Percutaneous sacroiliac joint screw is an important method for the treatment of pelvic fracture. Due to the complex anatomical structure around the sacroiliac joint, the misplacement of the sacroiliac screw is highly possible to damage the blood vessels and nerves. Surgical robot is more and more used in the operation of percutaneous sacroiliac joint screw because of its high accuracy, good stability and highsafety. At present, however, in robot assisted sacroiliac joint screw surgery, robots can only solve the problem of precise positioning while the surgical planning still relies on the doctor to do it manually. The purpose of this study is to implement an automatic, quantitative and optimized sacroiliac screw channel planning method based on the digital modeling and geometric calculation. Methods The pelvic CT scan data of 14 patients after sacroiliac screw surgery were processed into Mimics 17. 0 for the 3 D reconstruction of the pelvis. The pelvic models were used for automatic planning. The postoperative screws ' locations were calculated as manual planning results into the control group. A pelvic model was imported into the automatic planning module on which the starting point and end point of the screw channel were manually specified as two seed points. Two point arrays were automatically generated from two seed points. One starting point and one end point forms a screw channel. The channels completely located in the bone structure were selected as safe channels through the method of Boolean operation of volume models. The shortest distance between the surface points around the screw channel and the center line of the screw channel was defined as the security safety of the channel. The channel with the highest security degree was identified as the optimal channel. Results In 10 cases of S1 sacroiliac screw operation, automatic planning result was 6. 339 mm±1. 499 mm, better than manual planning result which was 2. 482 mm ± 1. 571 mm, and the difference was statistically significant. In 10 cases of S2 sacroiliac screw operation, automatic planning result was 5. 777 mm±1. 250 mm, better than manual planning result which was 1. 784 mm±1. 531 mm, and the difference was statistically significant. The qualification rate of S1 sacroiliac screw automatic planning result was 100 percent, better than the qualification rate of S1 sacroiliac screw manual planning result which was 80 percent. The qualification rate of S2 sacroiliac screw automatic planning result was 100 percent, better than the qualification rate of S2 sacroiliac screw manual planning result which was 60 percent. Conclusions The experimental results prove that the automatic planning method of sacroiliac screw in this study can realize the optimal channel planning of the individual sacroiliac screw, and the automatic planning result is better than the manual planning result. Our research provides a new idea for the study of automatic planning method of orthopaedic surgery.