2000 character limit reached
RuleFuser: An Evidential Bayes Approach for Rule Injection in Imitation Learned Planners and Predictors for Robustness under Distribution Shifts (2405.11139v3)
Published 18 May 2024 in cs.RO, cs.AI, and cs.LG
Abstract: 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.
- Posterior network: Uncertainty estimation without OOD samples via density-based pseudo-counts. Advances in Neural Information Processing Systems, 33:1356–1367, 2020.
- Multi-predictor fusion: Combining learning-based and rule-based trajectory predictors. arXiv preprint arXiv:2307.01408, 2023.
- nuPlan: A closed-loop ml-based planning benchmark for autonomous vehicles. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition ADP3 workshop, 2021.
- Driving everywhere with large language model policy adaptation. arXiv preprint arXiv:2402.05932, 2024.
- Robust traffic rules and knowledge representation for conflict resolution in autonomous driving. In Proc. of the 16th International Rule Challenge and 6th Doctoral Consortium @ RuleML+RR, 2022.
- Legal compliance checking of autonomous driving with formalized traffic rule exceptions. In Proc. of the Workshop on Logic Programming and Legal Reasoning in conjunction with 39th International Conference on Logic Programming (ICLP), 2022.
- Formalization of interstate traffic rules in temporal logic. In Proc. IEEE Intelligent Vehicles Symposium, pages 752–759. IEEE, 2020.
- Formalizing traffic rules for machine interpretability. In Proc. of IEEE Connected and Automated Vehicles Symposium (CAVS), pages 1–7. IEEE, 2020.
- Intention-aware motion planning with road rules. In Proc. of IEEE Int. Conf. on Automation Science and Engineering, pages 526–532. IEEE, 2020.
- Mining road traffic rules with signal temporal logic and grammar-based genetic programming. Applied Sciences, 11(22):10573, 2021.
- Encoding and monitoring responsibility sensitive safety rules for automated vehicles in signal temporal logic. In Proc. of the ACM-IEEE Int. Conf. on Formal Methods and Models for System Design, pages 1–11, 2019.
- Receding horizon planning with rule hierarchies for autonomous vehicles. arXiv preprint arXiv:2212.03323, 2022.
- Liability, ethics, and culture-aware behavior specification using rulebooks. In Proc. IEEE Conf. on Robotics and Automation, pages 8536–8542, 2019.
- Minimum-violation scltl motion planning for mobility-on-demand. In Proc. IEEE Conf. on Robotics and Automation, pages 1481–1488. IEEE, 2017.
- Minimum-violation ltl planning with conflicting specifications. In Proc. American Control Conference, pages 200–205, 2013.
- Social force model for pedestrian dynamics. Physical review E, 51(5):4282, 1995.
- Dsdnet: Deep structured self-driving network. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXI 16, pages 156–172. Springer, 2020.
- End-to-end interpretable neural motion planner. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pages 8660–8669, 2019.
- Combining neural networks and tree search for task and motion planning in challenging environments. In Proc. IEEE/RSJ Int. Conf. on Intelligent Robots & Systems, pages 6059–6066. IEEE, 2017.
- Road traffic law adaptive decision-making for self-driving vehicles. In Proc. IEEE Int. Conf. on Intelligent Transportation Systems, pages 2034–2041. IEEE, 2022.
- Follow the rules: Online signal temporal logic tree search for guided imitation learning in stochastic domains. In Proc. IEEE Conf. on Robotics and Automation, pages 1320–1326. IEEE, 2023.
- Trajectron++: Dynamically-feasible trajectory forecasting with heterogeneous data. In Proc. European Conf. on Computer Vision, pages 683–700. Springer, 2020.
- Scept: Scene-consistent, policy-based trajectory predictions for planning. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pages 17103–17112, 2022.
- Dropout as a bayesian approximation: Representing model uncertainty in deep learning. arXiv:1506.02142, 2015.
- Simple and scalable predictive uncertainty estimation using deep ensembles. Conf. on Neural Information Processing Systems, 2017.
- Sketching curvature for efficient out-of-distribution detection for deep neural networks. arXiv preprint arXiv:2102.12567, 2021.
- A scalable laplace approximation for neural networks. Int. Conf. on Learning Representations, 2018.
- Weight uncertainty in neural network. Int. Conf. on Machine Learning, 2015.
- David JC MacKay. A practical bayesian framework for backpropagation networks. Neural computation, 4(3), 1992.
- A simple unified framework for detecting out-of-distribution samples and adversarial attacks. Conf. on Neural Information Processing Systems, 2018.
- Ev Zisselman and Aviv Tamar. Deep residual flow for out of distribution detection. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2020.
- Do deep generative models know what they don’t know? Int. Conf. on Learning Representations, 2019.
- Uncertainty estimation using a single deep deterministic neural network. In Hal Daumé III and Aarti Singh, editors, Proceedings of the 37th International Conference on Machine Learning, volume 119 of Proceedings of Machine Learning Research, pages 9690–9700. PMLR, 13–18 Jul 2020.
- Regression prior networks. arXiv preprint arXiv:2006.11590, 2020.
- Predictive uncertainty estimation via prior networks. Conf. on Neural Information Processing Systems, 31, 2018.
- Interpretable self-aware neural networks for robust trajectory prediction. In Proc. Conf. on Robot Learning, pages 606–617. PMLR, 2023.
- Multimodal probabilistic model-based planning for human-robot interaction. In Proc. IEEE Conf. on Robotics and Automation, pages 3399–3406, 2018.
- Monitoring temporal properties of continuous signals. In Yassine Lakhnech and Sergio Yovine, editors, Formal Techniques, Modelling and Analysis of Timed and Fault-Tolerant Systems, pages 152–166. Springer Berlin Heidelberg, 2004.
- Back-propagation through signal temporal logic specifications: Infusing logical structure into gradient-based methods. arXiv preprint arXiv:2008.00097, 2020.
- The reasonable crowd: Towards evidence-based and interpretable models of driving behavior. In Proc. IEEE/RSJ Int. Conf. on Intelligent Robots & Systems, pages 6708–6715, 2021.
- Categorical traffic transformer: Interpretable and diverse behavior prediction with tokenized latent. arXiv preprint arXiv:2311.18307, 2023.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.
