Scene-Aware Explainable Multimodal Trajectory Prediction

2024-10-22

AI Summary

To address the critical limitations of existing trajectory prediction methods in complex traffic scenarios—namely, insufficient multi-agent joint reasoning and lack of interpretability—this paper proposes an interpretable multimodal prediction framework that jointly models scene semantics and agent interactions. Methodologically, it integrates a conditional diffusion model with an enhanced Shapley value attribution mechanism to achieve high-accuracy multimodal trajectory generation while enabling environment-factor-level interpretability. A scene-aware graph neural network captures dynamic agent interactions, and a dedicated multimodal decoder enhances output diversity. Evaluated on the Waymo Open Motion dataset, the method significantly outperforms state-of-the-art baselines in prediction accuracy (e.g., minADE/minFDE). Crucially, it identifies human-intuitive attribution factors—including lane topology and neighboring vehicles’ motion patterns—with high fidelity. This work establishes a new paradigm for autonomous driving decision-making that simultaneously advances performance and trustworthiness.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Multi-modal Vision
• Multiagent Systems: Adversarial Agents

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Advancements in intelligent technologies have significantly improved navigation in complex traffic environments by enhancing environment perception and trajectory prediction for automated vehicles. However, current research often overlooks the joint reasoning of scenario agents and lacks explainability in trajectory prediction models, limiting their practical use in real-world situations. To address this, we introduce the Explainable Conditional Diffusion-based Multimodal Trajectory Prediction (DMTP) model, which is designed to elucidate the environmental factors influencing predictions and reveal the underlying mechanisms. Our model integrates a modified conditional diffusion approach to capture multimodal trajectory patterns and employs a revised Shapley Value model to assess the significance of global and scenario-specific features. Experiments using the Waymo Open Motion Dataset demonstrate that our explainable model excels in identifying critical inputs and significantly outperforms baseline models in accuracy. Moreover, the factors identified align with the human driving experience, underscoring the model's effectiveness in learning accurate predictions. Code is available in our open-source repository: https://github.com/ocean-luna/Explainable-Prediction.

Problem

• Enhances explainability in trajectory prediction models
• Integrates multimodal trajectory patterns for accurate predictions
• Identifies critical environmental factors influencing predictions

Innovation

• Explainable Conditional Diffusion-based Multimodal Trajectory Prediction
• Modified conditional diffusion captures multimodal trajectory patterns
• Revised Shapley Value assesses global and scenario-specific feature significance