SocialMOIF: Multi-Order Intention Fusion for Pedestrian Trajectory Prediction

2025-04-22

AI Summary

Pedestrian short-term trajectory prediction faces two key challenges: high uncertainty in agent intent and complex, higher-order social interactions among nearby groups. To address these, we propose a multi-order intent fusion framework that unifies modeling of first-order (direct) and higher-order (indirect) social intent interactions—emphasizing the dominance of first-order interactions while explicitly capturing group-level intent propagation. Methodologically, we employ graph neural networks for multi-order intent encoding, design a trajectory distribution approximator and a global trajectory optimizer, and introduce a joint distance-direction geometric-aware loss function. Our approach achieves state-of-the-art performance across multiple dynamic and static benchmark datasets, significantly improving prediction accuracy, robustness, and physical plausibility. Moreover, it enhances model interpretability by explicitly disentangling intent hierarchies and interaction pathways.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The analysis and prediction of agent trajectories are crucial for decision-making processes in intelligent systems, with precise short-term trajectory forecasting being highly significant across a range of applications. Agents and their social interactions have been quantified and modeled by researchers from various perspectives; however, substantial limitations exist in the current work due to the inherent high uncertainty of agent intentions and the complex higher-order influences among neighboring groups. SocialMOIF is proposed to tackle these challenges, concentrating on the higher-order intention interactions among neighboring groups while reinforcing the primary role of first-order intention interactions between neighbors and the target agent. This method develops a multi-order intention fusion model to achieve a more comprehensive understanding of both direct and indirect intention information. Within SocialMOIF, a trajectory distribution approximator is designed to guide the trajectories toward values that align more closely with the actual data, thereby enhancing model interpretability. Furthermore, a global trajectory optimizer is introduced to enable more accurate and efficient parallel predictions. By incorporating a novel loss function that accounts for distance and direction during training, experimental results demonstrate that the model outperforms previous state-of-the-art baselines across multiple metrics in both dynamic and static datasets.

Problem

• Predict pedestrian trajectories with multi-order intention fusion
• Address uncertainty in agent intentions and group influences
• Enhance interpretability and accuracy in trajectory forecasting

Innovation

• Multi-order intention fusion for trajectory prediction
• Trajectory distribution approximator enhances interpretability
• Global trajectory optimizer enables parallel predictions