mlsanityData to 14 Oct 2026

RuleFuser: An Evidential Bayes Approach for Rule Injection in Imitation Learned Planners and Predictors for Robustness under Distribution Shifts

arXiv 2405.11139 ↗First posted 18 May 2024cs.RO cs.AI cs.LGcited by 2

On the map: Offline RL and imitation learning from data → Imitation and offline RL for driving

Abstract (arXiv)

Modern motion planners for autonomous driving frequently use imitation learning (IL) to draw from expert driving logs. Although IL benefits from its ability to glean nuanced and multi-modal human driving behaviors from large datasets, the resulting planners often struggle with out-of-distribution (OOD) scenarios and with traffic rule compliance. On the other hand, classical rule-based planners, by design, can generate safe traffic rule compliant behaviors while being robust to OOD scenarios, but these planners fail to capture nuances in agent-to-agent interactions and human drivers' intent. RuleFuser, an evidential framework, combines IL planners with classical rule-based planners to draw on the complementary benefits of both, thereby striking a balance between imitation and safety. Our approach, tested on the real-world nuPlan dataset, combines the IL planner's high performance in in-distribution (ID) scenarios with the rule-based planners' enhanced safety in out-of-distribution (OOD) scenarios, achieving a 38.43% average improvement on safety metrics over the IL planner without much detriment to imitation metrics in OOD scenarios.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Imitation and offline RL for driving, one of 13 directions in Offline RL and imitation learning from data. Builds on 30 of its 43 references that are in the map. First posted on arXiv 18 May 2024.

Builds on

30 of 43 references.

Closest ancestors · 1

Same direction, newest first

  1. 2022Imitation Is Not Enough: Robustifying Imitation with Reinforcement Learning for Challenging Driving Scenarios21 Dec

Other references · 29

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  2. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  3. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,926
  4. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  5. 2022Planning-oriented Autonomous Drivingcited by 1,087
  6. 2018Predictive Uncertainty Estimation via Prior Networkscited by 557
  7. 2021NuPlan: A closed-loop ML-based planning benchmark for autonomous vehiclescited by 493
  8. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  9. 2022Wayformer: Motion Forecasting via Simple & Efficient Attention Networkscited by 280
  10. 2022Motion Transformer with Global Intention Localization and Local Movement Refinementcited by 270
  11. 2021End-to-end Interpretable Neural Motion Plannercited by 242
  12. 2021Scene Transformer: A unified architecture for predicting multiple agent trajectoriescited by 165
17 more
  1. 2020Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Countscited by 164
  2. 2023The Safety Filter: A Unified View of Safety-Critical Control in Autonomous Systemscited by 88
  3. 2020DSDNet: Deep Structured self-Driving Networkcited by 62
  4. 2020Backpropagation through Signal Temporal Logic Specifications: Infusing Logical Structure into Gradient-Based Methodscited by 54
  5. 2024Driving Everywhere with Large Language Model Policy Adaptationcited by 53
  6. 2017Multimodal Probabilistic Model-Based Planning for Human-Robot Interactioncited by 50
  7. 2019Liability, Ethics, and Culture-Aware Behavior Specification using Rulebookscited by 43
  8. 2021Sketching Curvature for Efficient Out-of-Distribution Detection for Deep Neural Networkscited by 32
  9. 2020Regression Prior Networkscited by 28
  10. 2022Interpretable Self-Aware Neural Networks for Robust Trajectory Predictioncited by 22
  11. 2017Combining Neural Networks and Tree Search for Task and Motion Planning in Challenging Environmentscited by 19
  12. 2020Formalizing Traffic Rules for Machine Interpretabilitycited by 13
  13. 2021The Reasonable Crowd: Towards evidence-based and interpretable models of driving behaviorcited by 12
  14. 2013Minimum-violation LTL Planning with Conflicting Specificationscited by 7
  15. 2022Follow The Rules: Online Signal Temporal Logic Tree Search for Guided Imitation Learning in Stochastic Domainscited by 7
  16. 2022Receding Horizon Planning with Rule Hierarchies for Autonomous Vehiclescited by 7
  17. 2023Multi-Predictor Fusion: Combining Learning-based and Rule-based Trajectory Predictorscited by 3

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 1.

2 papers.

  1. 2024Adaptive Prediction Ensemble: Improving Out-of-Distribution Generalization of Motion Forecastingcited by 7
  2. 2025Trends in Motion Prediction Toward Deployable and Generalizable Autonomy: A Revisit and Perspectivescited by 5

Similar, not cited

Close in wording, no citation either way

  1. 2026Learning to Explain While Planning: Rule-Aligned Diffusion Planning for Autonomous Driving
  2. 2024PLUTO: Pushing the Limit of Imitation Learning-based Planning for Autonomous Driving
  3. 2025Sequence of Expert: Boosting Imitation Planners for Autonomous Driving through Temporal Alternation
  4. 2023Rethinking Imitation-based Planner for Autonomous Driving
  5. 2022Learning from Demonstrations of Critical Driving Behaviours Using Driver's Risk Field
  6. 2024Validity Learning on Failures: Mitigating the Distribution Shift in Autonomous Vehicle Planning
  7. 2026Physics-Grounded Causal Auditing of End-to-End Driving Planners
  8. 2022Embedding Synthetic Off-Policy Experience for Autonomous Driving via Zero-Shot Curricula