Achieving accurate implant positioning and optimal gap balance is essential in total knee arthroplasty (TKA), yet conventional instrumentation is associated with a high rate of alignment errors. Although neutral mechanical alignment has traditionally been the target, increasing attention is being directed toward more anatomical alignment strategies to improve functional outcomes and reduce the need for extensive soft-tissue releases. These limitations have driven the adoption of advanced technologies such as computer-assisted and robotic-assisted surgery.

Robotic-assisted TKA has evolved considerably over the past two decades, often relying on preoperative imaging and autonomous cutting tools, which may enhance accuracy but at the cost of increased complexity, setup time, and expense. The Robin robotic system introduces a novel, image-free, collaborative approach, in which a robotic arm positions cutting jigs while the surgeon maintains full control of bone resections, preserving tactile feedback and the workflow of conventional instrumentation. Intraoperative landmark registration is performed manually using rigid body trackers attached to the femur and tibia, followed by real-time planning of resection planes. Consequently, the system’s accuracy is highly dependent on the precision and reproducibility of landmark registration.

This study aims to evaluate the accuracy of bone resections performed with the Robin robotic system and to assess the reproducibility of landmark registration among different users. Secondary objectives include the analysis of surgical time efficiency and inter-operator consistency.