The rapid development of large language models (LLMs) and multimodal large models (MLMs) has introduced transformative opportunities for autonomous driving systems. These advanced models provide robust support for the realization of more intelligent, safer, and efficient autonomous driving. In this paper, we present a systematic review on the integration of LLMs and MLMs in autonomous driving systems. First, we provide an overview of the evolution of LLMs and MLMs, along with a detailed analysis of the architecture of autonomous driving systems. Next, we explore the applications of LLMs and MLMs in key components such as perception, prediction, decision making, planning, multitask processing, and human–machine interaction. Additionally, this paper reviews the core technologies involved in integrating LLMs and MLMs with autonomous driving systems, including multimodal fusion, knowledge distillation, prompt engineering, and supervised fine tuning. Finally, we provide an in-depth analysis of the major challenges faced by autonomous driving systems powered by large models, offering new perspectives for future research. Compared to existing review articles, this paper not only systematically examines the specific applications of LLMs and MLMs in autonomous driving systems but also delves into the key technologies and potential challenges involved in their integration. By comprehensively organizing and analyzing the current literature, this review highlights the application potential of large models in autonomous driving and offers insights and recommendations for improving system safety and efficiency.
Fatigue can impair attention, reaction ability, and judgment, making it a critical safety concern in various high-risk scenarios. Due to its advantages of non-contact operation, low cost, and easy deployment, computer vision-based fatigue detection has become a major research focus in this field. This review systematically summarizes the technological evolution of fatigue detection, from traditional machine learning and deep learning to LLMs. It expands the application scenarios from driving to aviation, industrial, medical, educational, and sports domains, and provides comparative analyses of eight classical fatigue detection datasets and representative algorithms. Furthermore, five key challenges are identified, including environmental interference, generalization capability, individual differences, privacy and ethical issues, and the practical deployment of LLMs, followed by discussions on future development directions. These efforts systematically address the limitations of existing reviews in terms of application scenario coverage, frontier technology analysis, and comparisons of datasets and algorithms. The findings indicate that fatigue detection technology is evolving from task-specific visual models toward general intelligent understanding paradigms. The introduction of LLMs is expected to overcome the limitations of conventional methods in terms of generalization and interpretability, thereby laying a foundation for the large-scale application of fatigue detection technologies.
Fatigue can impair attention, reaction ability, and judgment, making it a critical safety concern in various high-risk scenarios. Due to its advantages of non-contact operation, low cost, and easy deployment, computer vision-based fatigue detection has become a major research focus in this field. This review systematically summarizes the technological evolution of fatigue detection, from traditional machine learning and deep learning to LLMs. It expands the application scenarios from driving to aviation, industrial, medical, educational, and sports domains, and provides comparative analyses of eight classical fatigue detection datasets and representative algorithms. Furthermore, five key challenges are identified, including environmental interference, generalization capability, individual differences, privacy and ethical issues, and the practical deployment of LLMs, followed by discussions on future development directions. These efforts systematically address the limitations of existing reviews in terms of application scenario coverage, frontier technology analysis, and comparisons of datasets and algorithms. The findings indicate that fatigue detection technology is evolving from task-specific visual models toward general intelligent understanding paradigms. The introduction of LLMs is expected to overcome the limitations of conventional methods in terms of generalization and interpretability, thereby laying a foundation for the large-scale application of fatigue detection technologies.