mlsanityData to 14 Oct 2026

AdvDO: Realistic Adversarial Attacks for Trajectory Prediction

arXiv 2209.08744 ↗First posted 19 Sep 2022cs.LG cs.AI cs.CRcited by 49

Not on the map. Why?

Abstract (arXiv)

Trajectory prediction is essential for autonomous vehicles (AVs) to plan correct and safe driving behaviors. While many prior works aim to achieve higher prediction accuracy, few study the adversarial robustness of their methods. To bridge this gap, we propose to study the adversarial robustness of data-driven trajectory prediction systems. We devise an optimization-based adversarial attack framework that leverages a carefully-designed differentiable dynamic model to generate realistic adversarial trajectories. Empirically, we benchmark the adversarial robustness of state-of-the-art prediction models and show that our attack increases the prediction error for both general metrics and planning-aware metrics by more than 50% and 37%. We also show that our attack can lead an AV to drive off road or collide into other vehicles in simulation. Finally, we demonstrate how to mitigate the adversarial attacks using an adversarial training scheme.

Key facts. Cited by 49 papers in the map (data through 14 Oct 2026). Builds on 31 of its 46 references that are in the map. First posted on arXiv 19 Sep 2022.

Builds on

31 of 46 references.

References · 31

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  3. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  4. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  5. 2020Fast is better than free: Revisiting adversarial trainingcited by 727
  6. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 678
  7. 2019Adversarial Training for Free!cited by 643
  8. 2019On Evaluating Adversarial Robustnesscited by 447
  9. 2021AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecastingcited by 331
  10. 2017Adversarial Examples for Semantic Segmentation and Object Detectioncited by 275
  11. 2017MagNet: a Two-Pronged Defense against Adversarial Examplescited by 269
  12. 2018Generating Adversarial Examples with Adversarial Networkscited by 258
19 more
  1. 2019Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networkscited by 229
  2. 2019Towards Stable and Efficient Training of Verifiably Robust Neural Networkscited by 205
  3. 2021TrafficSim: Learning to Simulate Realistic Multi-Agent Behaviorscited by 193
  4. 2019PRECOG: PREdiction Conditioned On Goals in Visual Multi-Agent Settingscited by 179
  5. 2018Generating 3D Adversarial Point Cloudscited by 144
  6. 2021Generating Useful Accident-Prone Driving Scenarios via a Learned Traffic Priorcited by 139
  7. 2021AdvSim: Generating Safety-Critical Scenarios for Self-Driving Vehiclescited by 137
  8. 2021Multimodal Trajectory Prediction Conditioned on Lane-Graph Traversalscited by 113
  9. 2019Universal Physical Camouflage Attacks on Object Detectorscited by 91
  10. 2022On Adversarial Robustness of Trajectory Prediction for Autonomous Vehiclescited by 90
  11. 2019AdvPC: Transferable Adversarial Perturbations on 3D Point Cloudscited by 84
  12. 2020Learning to Collide: An Adaptive Safety-Critical Scenarios Generating Methodcited by 75
  13. 2020PatchAttack: A Black-box Texture-based Attack with Reinforcement Learningcited by 67
  14. 2019ME-Net: Towards Effective Adversarial Robustness with Matrix Estimationcited by 65
  15. 2020Multimodal Safety-Critical Scenarios Generation for Decision-Making Algorithms Evaluationcited by 55
  16. 2021Injecting Planning-Awareness into Prediction and Detection Evaluationcited by 21
  17. 2018Thwarting Adversarial Examples: An L0-RobustSparse Fourier Transformcited by 13
  18. 2019Efficient Autonomy Validation in Simulation with Adaptive Stress Testingcited by 13
  19. 2019Geometry-Aware Generation of Adversarial Point Cloudscited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 3; 2023: 17; 2024: 14; 2025: 7; 2026: 8.

49 papers.

Sort:

By area:

  1. 2024UniTraj: A Unified Framework for Scalable Vehicle Trajectory Predictioncited by 52
  2. 2023Uncovering the Missing Pattern: Unified Framework Towards Trajectory Imputation and Predictioncited by 47
  3. 2023RealGen: Retrieval Augmented Generation for Controllable Traffic Scenarioscited by 46
  4. 2023Traj-MAE: Masked Autoencoders for Trajectory Predictioncited by 42
  5. 2023SAFE-SIM: Safety-Critical Closed-Loop Traffic Simulation with Diffusion-Controllable Adversariescited by 38
  6. 2022Behavioral Intention Prediction in Driving Scenes: A Surveycited by 30
  7. 2023Defending against Adversarial Audio via Diffusion Modelcited by 23
  8. 2023Reinforcement Learning with Human Feedback for Realistic Traffic Simulationcited by 21
  9. 2023Joint Metrics Matter: A Better Standard for Trajectory Forecastingcited by 17
  10. 2023Action-slot: Visual Action-centric Representations for Multi-label Atomic Activity Recognition in Traffic Scenescited by 12
  11. 2022Targeted Adversarial Attacks against Neural Network Trajectory Predictorscited by 11
  12. 2024CaDRE: Controllable and Diverse Generation of Safety-Critical Driving Scenarios using Real-World Trajectoriescited by 11
37 more
  1. 2023Adv3D: Generating Safety-Critical 3D Objects through Closed-Loop Simulationcited by 10
  2. 2024SEAL: Towards Safe Autonomous Driving via Skill-Enabled Adversary Learning for Closed-Loop Scenario Generationcited by 10
  3. 2023TrajPAC: Towards Robustness Verification of Pedestrian Trajectory Prediction Modelscited by 8
  4. 2023Adversarial Backdoor Attack by Naturalistic Data Poisoning on Trajectory Prediction in Autonomous Drivingcited by 6
  5. 2024Towards Deviation-Robust Agent Navigation via Perturbation-Aware Contrastive Learningcited by 6
  6. 2024Certified Human Trajectory Predictioncited by 6
  7. 2022Benchmark for Models Predicting Human Behavior in Gap Acceptance Scenarioscited by 5
  8. 2025Adversarial Generation and Collaborative Evolution of Safety-Critical Scenarios for Autonomous Vehiclescited by 5
  9. 2023Sim-to-Real Causal Transfer: A Metric Learning Approach to Causally-Aware Interaction Representationscited by 4
  10. 2024Catastrophic Overfitting: A Potential Blessing in Disguisecited by 4
  11. 2024Decictor: Towards Evaluating the Robustness of Decision-Making in Autonomous Driving Systemscited by 4
  12. 2024Module-wise Adaptive Adversarial Training for End-to-end Autonomous Drivingcited by 4
  13. 2025Temporal Logic-Based Multi-Vehicle Backdoor Attacks against Offline RL Agents in End-to-end Autonomous Drivingcited by 4
  14. 2023Sparse Pedestrian Character Learning for Trajectory Predictioncited by 3
  15. 2024A First Physical-World Trajectory Prediction Attack via LiDAR-induced Deceptions in Autonomous Drivingcited by 3
  16. 2024Evaluating Human Trajectory Prediction with Metamorphic Testingcited by 3
  17. 2024Adversarial and Reactive Traffic Entities for Behavior-Realistic Driving Simulation: A Reviewcited by 3
  18. 2024RealTraj: Towards Real-World Pedestrian Trajectory Forecastingcited by 3
  19. 2025ATARS: An Aerial Traffic Atomic Activity Recognition and Temporal Segmentation Datasetcited by 2
  20. 2025Safety2Drive: Safety-Critical Scenario Benchmark for the Evaluation of Autonomous Drivingcited by 2
  21. 2025STEP: Structured Training and Evaluation Platform for benchmarking trajectory prediction modelscited by 2
  22. 2023Robustness Benchmark of Road User Trajectory Prediction Models for Automated Drivingcited by 1
  23. 2023Flying Adversarial Patches: Manipulating the Behavior of Deep Learning-based Autonomous Multirotorscited by 1
  24. 2023Kidnapping Deep Learning-based Multirotors using Optimized Flying Adversarial Patchescited by 1
  25. 2024SA-Attack: Speed-adaptive stealthy adversarial attack on trajectory predictioncited by 1
  26. 2025Towards Benchmarking and Assessing the Safety and Robustness of Autonomous Driving on Safety-critical Scenarioscited by 1
  27. 2025Adversarial Examples in Environment Perception for Automated Driving (Review)cited by 1
  28. 2026Evaluating randomized smoothing as a defense against adversarial attacks in trajectory predictioncited by 1
  29. 2026Mitigating Error Amplification in Fast Adversarial Trainingcited by 1
  30. 2023Manipulating Trajectory Prediction with Backdoorscited by 0
  31. 2024Hacking Predictors Means Hacking Cars: Using Sensitivity Analysis to Identify Trajectory Prediction Vulnerabilities for Autonomous Driving Securitycited by 0
  32. 2026Super Agents and Confounders: Influence of surrounding agents on vehicle trajectory predictioncited by 0
  33. 2026Unveiling the Backdoor Mechanism Hidden Behind Catastrophic Overfitting in Fast Adversarial Trainingcited by 0
  34. 2026EvoDrive: Pareto Evolution for Safety-Critical Autonomous Driving via Self-Improving LLM Agentscited by 0
  35. 2026TrajRS: Towards Certified Robustness in Pedestrian Trajectory Predictioncited by 0
  36. 2026Collision Snapshot Guided Time-Reversed Safety-Critical Scenario Generationcited by 0
  37. 2026Action-Slot: Structured Action-Centric Representation Learning for Multi-Agent Atomic Activity Understandingcited by 0