Obtaining real-time pedestrian trajectory information in the driving direction is critical to ensuring the safety of autonomous vehicles. In this study, a joint behavior prediction model, which is based on the YOLOv5n pedestrian detection, binocular vision, and Kalman filtering, is proposed to address the environmental perception challenge of avoiding pedestrian collisions during autonomous driving. Depth information is obtained through a binocular camera, and the YOLOv5n model is utilized to detect and track pedestrians. Meanwhile, the adaptive Kalman filtering algorithm based on Particle Swarm Optimization (PSO) is utilized to estimate and predict the dynamic trajectories of pedestrians. The primary goal of this research is to improve the accuracy and reliability of pedestrian trajectory prediction in autonomous driving systems. Experiments demonstrate that the proposed model excels in complex traffic environments and effectively enhances the safety of driving. This study offers significant technical support for autonomous vehicles to make precise decisions in dynamic environments.
A pedestrian trajectory prediction method in intelligent driving
Second International Conference on Image Processing and Artificial Intelligence (ICIPAI 2025) ; 2025 ; Changchun, China
Proc. SPIE ; 13780 ; 137800L
2025-08-11
Conference paper
Electronic Resource
English
Pedestrian trajectory prediction method and automatic driving method of vehicle
European Patent Office | 2023
|Evaluating Pedestrian Trajectory Prediction Methods With Respect to Autonomous Driving
IEEE | 2024
|