This paper aims to develop a dynamic trust prediction model for teleoperated robotic systems in human-robot interaction (HRI) scenarios. Our model captures trust fluctuations influenced by task performance, cognitive load, and physiological responses. In our experiments, participants teleoperated a robotic arm under three conditions: without guidance (C1), with verbal guidance (C2), and with a combination of verbal and visual guidance (C3). Trust levels were measured after each condition, and cognitive load was assessed using the NASA TLX and physiological sensors. Our dynamic Bayesian network model demonstrated significant improvements in predictive accuracy, achieving 89% accuracy, 91% precision, 92.5% recall, and an 84% F1 score. The results indicated significant variations in trust levels across conditions (p < 0.032) and an inverse relationship between cognitive load and trust (r = –0.632, p < 0.01). The model effectively captured the dynamic interplay between trust and cognitive load, highlighting the importance of adaptive system design to maintain high trust levels. By leveraging physiological indicators and performance metrics, the dynamic trust model provides a nuanced understanding of trust evolution, facilitating better human-robot collaboration.