TrajMamba: An Ego-Motion-Guided Mamba Model for Pedestrian Trajectory Prediction from an Egocentric Perspective

Researchers propose TrajMamba, a new AI model that uses Mamba architecture to predict pedestrian movement from an ego-centric perspective for autonomous driving applications. The model integrates pedestrian motion and ego-vehicle movement data to achieve state-of-the-art performance on PIE and JAAD datasets.

• →TrajMamba introduces a novel Mamba-based approach for predicting pedestrian trajectories from an egocentric camera perspective.
• →The model uses dual Mamba encoders to separately process pedestrian motion and ego-vehicle movement features.
• →An ego-motion guided decoder integrates both motion types to capture complex relative movement dynamics.
• →The approach achieves state-of-the-art performance on standard PIE and JAAD pedestrian prediction datasets.
• →This research advances AI capabilities for autonomous driving and robot navigation safety systems.