Objective: To evaluate the accuracy and efficiency of an autonomous dental robotic system in performing calcified canal localization and negotiation on standardized three-dimensional (3D)-printed jaw models. Methods: Two pairs of standardized 3D-printed maxillary and mandibular models, containing a total of 56 teeth, 92 calcified root canals, were used. Virtual access paths were planned based on preoperative cone beam CT (CBCT) and intraoral scan data using digital dental design software. An experienced operator controlled the autonomous dental robotic system to perform canal localization, and drilling time were recorded. Postoperative CBCT images were registered with preoperative plans to calculate accuracy parameters, including coronal deviation, apical deviation, and angular deviation. The residual thickness of canal wall at drilling end level, and the success rate of calcified canal location was recorded as well. Multiple linear regression analysis was performed to evaluate the effects of tooth position, jaw position, operation side and calcification depth on accuracy and drilling time. Results: The coronal deviation, apical deviation, and angular deviation of the calcified canal negotiation assisted by the autonomous dental robot were 0.35 (0.21) mm, 0.47 (0.27) mm, and 1.17° (1.35°), respectively, with an average drilling time of 39.00 (25.25) s. The residual dentine thickness of canal wall at drilling end level was (1.24±0.51) mm. The success rate of calcified canal location was 95.7% (88/92). Multiple linear regression analysis revealed that jaw position had a significant effect on coronal deviation (P0.05). Moreover, calcification depth had a significant effect on drilling time (P0.05). Conclusions: The autonomous dental robotic system demonstrated high accuracy and efficiency in calcified canal localization in vitro.