Context-aware multi-task learning for pedestrian intent and trajectory prediction
Farzeen Munir, Tomasz Piotr Kucner
Finnish Center for Artificial Intelligence Aalto University
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
The advancement of socially-aware autonomous vehicles hinges on precise modeling of human behavior. Within this broad paradigm, the specific challenge lies in accurately predicting pedestrian’s trajectory and intention. Traditional methodologies have leaned heavily on historical trajectory data, frequently overlooking vital contextual cues such as pedestrian-specific traits and environmental factors. Furthermore, there is a notable knowledge gap as trajectory and intention prediction have largely been approached as separate problems, despite their mutual dependence. To bridge this gap, we introduce PTINet (Pedestrian Trajectory and Intention Prediction Network), which jointly learns the trajectory and intention prediction by combining past trajectory observations, local contextual features (individual pedestrian behaviors), and global features (signs, markings etc.). The efficacy of our approach is evaluated on widely used public datasets: JAAD, PIE and TITAN, where it has demonstrated superior performance over existing state-of-the-art models in trajectory and intention prediction. The results from our experiments and ablation studies robustly validate PTINet’s effectiveness in jointly exploring intention and trajectory prediction for pedestrian behavior modeling. The experimental evaluation indicates the advantage of using global and local contextual features for pedestrian trajectory and intention prediction. The effectiveness of PTINet in predicting pedestrian behavior paves the way for the development of automated systems capable of seamlessly interacting with pedestrians in urban settings https://github.com/munirfarzeen/PTINet . • In this work, we asked whether predicting one aspect (intention) without the other (trajectory) limits the understanding of pedestrian behavior. Our results support this hypothesis by showing improved performance using our joint prediction approach. • We developed a novel multi-task framework, PTINet, that combines Local Contextual Features (LCF), such as pedestrian-specific attributes, with Global Features (GF)from image data and optical flow to analyze complex spatial–temporal patterns. • We chose a traditional LSTM over more complex transformers due to lower resource demands. Despite its simplicity, our approach outperforms transformer-based models in predicting pedestrian behavior. • Extensive experiments and ablation studies on benchmark datasets show PTINet’s superior performance, significantly outperforming state-of-the-art prediction models.
逐年被引趋势
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
工程Autonomous Vehicle Technology and Safety
Evacuation and Crowd Dynamics · Traffic and Road Safety