mlsanityData to 14 Oct 2026

Towards Explainable Motion Prediction using Heterogeneous Graph Representations

arXiv 2212.03806 ↗First posted 7 Dec 2022cs.AI cs.ROcited by 10

Not on the map. Why?

Abstract (arXiv)

Motion prediction systems aim to capture the future behavior of traffic scenarios enabling autonomous vehicles to perform safe and efficient planning. The evolution of these scenarios is highly uncertain and depends on the interactions of agents with static and dynamic objects in the scene. GNN-based approaches have recently gained attention as they are well suited to naturally model these interactions. However, one of the main challenges that remains unexplored is how to address the complexity and opacity of these models in order to deal with the transparency requirements for autonomous driving systems, which includes aspects such as interpretability and explainability. In this work, we aim to improve the explainability of motion prediction systems by using different approaches. First, we propose a new Explainable Heterogeneous Graph-based Policy (XHGP) model based on an heterograph representation of the traffic scene and lane-graph traversals, which learns interaction behaviors using object-level and type-level attention. This learned attention provides information about the most important agents and interactions in the scene. Second, we explore this same idea with the explanations provided by GNNExplainer. Third, we apply counterfactual reasoning to provide explanations of selected individual scenarios by exploring the sensitivity of the trained model to changes made to the input data, i.e., masking some elements of the scene, modifying trajectories, and adding or removing dynamic agents. The explainability analysis provided in this paper is a first step towards more transparent and reliable motion prediction systems, important from the perspective of the user, developers and regulatory agencies. The code to reproduce this work is publicly available at https://github.com/sancarlim/Explainable-MP/tree/v1.1.

Key facts. Cited by 10 papers in the map (data through 14 Oct 2026). Builds on 43 of its 63 references that are in the map. First posted on arXiv 7 Dec 2022.

Builds on

43 of 63 references.

References · 43

  1. 2017Attention Is All You Needcited by 60,361
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  3. 2017Graph Attention Networkscited by 8,923
  4. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  5. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  6. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
  7. 2019Attention is not Explanationcited by 929
  8. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 737
  9. 2020VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representationcited by 530
  10. 2018Explaining Explanations: An Overview of Interpretability of Machine Learningcited by 508
  11. 2020Parameterized Explainer for Graph Neural Networkcited by 426
  12. 2020Explainability in Graph Neural Networks: A Taxonomic Surveycited by 384
31 more
  1. 2020Learning Lane Graph Representations for Motion Forecastingcited by 382
  2. 2020TNT: Target-driveN Trajectory Predictioncited by 347
  3. 2021DenseTNT: End-to-end Trajectory Prediction from Dense Goal Setscited by 295
  4. 2020XGNN: Towards Model-Level Explanations of Graph Neural Networkscited by 211
  5. 2020PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networkscited by 211
  6. 2020GraphLIME: Local Interpretable Model Explanations for Graph Neural Networkscited by 209
  7. 2019Explainability Techniques for Graph Convolutional Networkscited by 161
  8. 2020Interpreting Graph Neural Networks for NLP With Differentiable Edge Maskingcited by 153
  9. 2020Higher-Order Explanations of Graph Neural Networks via Relevant Walkscited by 138
  10. 2021Generative Causal Explanations for Graph Neural Networkscited by 124
  11. 2021CF-GNNExplainer: Counterfactual Explanations for Graph Neural Networkscited by 119
  12. 2021LaneRCNN: Distributed Representations for Graph-Centric Motion Forecastingcited by 113
  13. 2021Multimodal Trajectory Prediction Conditioned on Lane-Graph Traversalscited by 113
  14. 2019Spatially-Aware Graph Neural Networks for Relational Behavior Forecasting from Sensor Datacited by 104
  15. 2021THOMAS: Trajectory Heatmap Output with learned Multi-Agent Samplingcited by 101
  16. 2022On Adversarial Robustness of Trajectory Prediction for Autonomous Vehiclescited by 90
  17. 2020What-If Motion Prediction for Autonomous Drivingcited by 80
  18. 2019GRIP++: Enhanced Graph-based Interaction-aware Trajectory Prediction for Autonomous Drivingcited by 73
  19. 2020Trajectory Forecasts in Unknown Environments Conditioned on Grid-Based Planscited by 72
  20. 2020Interpretable and Efficient Heterogeneous Graph Convolutional Networkcited by 48
  21. 2020Quantifying Explainers of Graph Neural Networks in Computational Pathologycited by 35
  22. 2019Layerwise Relevance Visualization in Convolutional Text Graph Classifierscited by 32
  23. 2020Probabilistic Multi-modal Trajectory Prediction with Lane Attention for Autonomous Vehiclescited by 28
  24. 2021Graph and Recurrent Neural Network-based Vehicle Trajectory Prediction For Highway Drivingcited by 19
  25. 2021Injecting Knowledge in Data-driven Vehicle Trajectory Predictorscited by 18
  26. 2020Towards Explainable Graph Representations in Digital Pathologycited by 17
  27. 2021SCOUT: Socially-COnsistent and UndersTandable Graph Attention Network for Trajectory Prediction of Vehicles and VRUscited by 16
  28. 2021Quantitative Evaluation of Explainable Graph Neural Networks for Molecular Property Predictioncited by 15
  29. 2022Testing predictive automated driving systems: lessons learned and future recommendationscited by 6
  30. 2022Towards trustworthy multi-modal motion prediction: Holistic evaluation and interpretability of outputscited by 4
  31. 2022Investigating Brain Connectivity with Graph Neural Networks and GNNExplainercited by 3

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 5; 2025: 4.

10 papers.

  1. 2023Counterfactual Learning on Graphs: A Surveycited by 22
  2. 2024Advancing Explainable Autonomous Vehicle Systems: A Comprehensive Review and Research Roadmapcited by 6
  3. 2024Testing autonomous vehicles and AI: perspectives and challenges from cybersecurity, transparency, robustness and fairnesscited by 3
  4. 2024Towards Data-centric Machine Learning on Directed Graphs: a Surveycited by 2
  5. 2025TPK: Trustworthy Trajectory Prediction Integrating Prior Knowledge For Interpretability and Kinematic Feasibilitycited by 1
  6. 2025SVBRD-LLM: Self-Verifying Behavioral Rule Discovery for Autonomous Vehicle Identificationcited by 1
  7. 2024Interpretable Responsibility Sharing as a Heuristic for Task and Motion Planningcited by 0
  8. 2024DiVR: incorporating context from diverse VR scenes for human trajectory predictioncited by 0
  9. 2025GC-GAT: Multimodal Vehicular Trajectory Prediction using Graph Goal Conditioning and Cross-context Attentioncited by 0
  10. 2025GContextFormer: A global context-aware hybrid multi-head attention approach with scaled additive aggregation for multimodal trajectory predictioncited by 0