Background <p>Limited training equipment and instructors at training centers cannot fulfill the simultaneous training required of numerous trainees. Optimizing the training organization can improve the training efficiency and trainee performance to enhance the overall training effectiveness. This study aims to evaluate the effectiveness and feasibility of group competition mechanism under peer evaluation during robot training.</p> Methods <p>A randomized controlled trial was conducted with 60 robotic-naive medical students, divided into two groups: independent training and team-based learning with a competitive element. Both groups performed Pick-and-Place (PP), Match-Board-2 (MB), and Thread-the-Rings-1 (TR) tasks on a dV-Trainer robot simulator. Training scores were automatically recorded to construct learning curves and calculate scoring efficiencies, which were compared between the two groups.</p> Results <p>Compared with the conventional training group, the group utilizing the group competition training format demonstrated significantly higher average operational scores, scoring efficiency, and learning curves (<i>P</i> &lt; 0.05). In addition, the group competition mode demonstrates a higher performance in cultivating the trainees’ satisfactory operational habits (<i>P</i> &lt; 0.001).</p> Conclusions <p>Based on task-reporting strategies guided by peer evaluation, the group competition mechanism is a novel and effective model for improving robot simulator training. This method holds promise as an effective tool for the surgical education of practicing surgeons and clinical medical students.</p>