SCoTT: Strategic Chain-of-Thought Tasking for Wireless-Aware Robot Navigation in Digital Twins

The introduction of SCoTT marks a significant advancement in robot navigation, particularly under wireless performance constraints. Traditional path planning methods often struggle with high computational costs when integrating such constraints. SCoTT addresses this by leveraging vision-language models to optimize both path gains and trajectory lengths using data from digital twins. In comparative studies, SCoTT demonstrated its effectiveness by achieving path gains within 2% of the optimal DP-WA* algorithm while consistently producing shorter trajectories. Additionally, SCoTT's design allows it to accelerate the DP-WA* algorithm by reducing its search space, leading to execution time savings of up to 62%. This dual focus on performance and efficiency positions SCoTT as a promising solution for future robotic navigation applications, particularly in complex environments where wireless communication is critical.