Researchers at the University of Georgia have developed a new artificial intelligence model to enhance self-driving car safety by predicting the movement of nearby traffic and planning safe vehicle movements. According to Qianwen Li, lead author of the study and an assistant professor in UGA's College of Engineering, the current approach used in self-driving cars can lead to crashes and near-misses due to discrepancies between predicted and actual traffic movements. The new model takes into account prediction errors and aims to balance safety and mobility. Li's group is also exploring the use of large language models like ChatGPT for self-driving car operations, but notes that traditional trajectory optimization models are better suited for planning specific trajectories. The study, published in Transportation Research Part E, was co-authored by Handong Yao, Xiaopeng Li, and Chenyang Yu, and has implications for companies developing self-driving car technology, including those using artificial intelligence to predict traffic movements and prevent accidents on the road.
↧