Can Autonomous Vehicles Identify, Recover From, and Adapt to Distribution Shifts?

Abstract

Out-of-training-distribution (OOD) scenarios are a common challenge of learning agents at deployment, typically leading to arbitrary deductions and poorly-informed decisions. In principle, detection of and adaptation to OOD scenes can mitigate their adverse effects. In this paper, we highlight the limitations of current approaches to novel driving scenes and propose an epistemic uncertainty-aware planning method, called \emph{robust imitative planning} (RIP). Our method can detect and recover from some distribution shifts, reducing the overconfident and catastrophic extrapolations in OOD scenes. If the model's uncertainty is too great to suggest a safe course of action, the model can instead query the expert driver for feedback, enabling sample-efficient online adaptation, a variant of our method we term \emph{adaptive robust imitative planning} (AdaRIP). Our methods outperform current state-of-the-art approaches in the nuScenes \emph{prediction} challenge, but since no benchmark evaluating OOD detection and adaption currently exists to assess \emph{control}, we introduce an autonomous car novel-scene benchmark, \texttt{CARNOVEL}, to evaluate the robustness of driving agents to a suite of tasks with distribution shifts.

arXiv commentThe first two authors contributed equally. Accepted at ICML 2020. Supplementary videos and code available at: https://sites.google.com/view/av-detect-recover-adapt

Cited by in corpus (19)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• A Survey of End-to-End Driving: Architectures and Training Methods
• A Survey of Zero-shot Generalisation in Deep Reinforcement Learning
• End-to-end Autonomous Driving with Semantic Depth Cloud Mapping and Multi-agent
• Causal Navigation by Continuous-time Neural Networks
• Perfect density models cannot guarantee anomaly detection
• Scaling Hamiltonian Monte Carlo Inference for Bayesian Neural Networks with Symmetric Splitting
• Scenario-Transferable Semantic Graph Reasoning for Interaction-Aware Probabilistic Prediction
• Constrained Model-based Reinforcement Learning with Robust Cross-Entropy Method
• Improving the Generalizability of Trajectory Prediction Models with Frenet-Based Domain Normalization
• ATOM: Robustifying Out-of-distribution Detection Using Outlier Mining
• Hybrid Imitative Planning with Geometric and Predictive Costs in Off-road Environments
• Shifts: A Dataset of Real Distributional Shift Across Multiple Large-Scale Tasks
• Explaining The Efficacy of Counterfactually Augmented Data
• CausalCity: Complex Simulations with Agency for Causal Discovery and Reasoning
• PsiPhi-Learning: Reinforcement Learning with Demonstrations using Successor Features and Inverse Temporal Difference Learning
• Watch out for the risky actors: Assessing risk in dynamic environments for safe driving
• 3rd Place Solution for NeurIPS 2021 Shifts Challenge: Vehicle Motion Prediction
• Counterfactual Maximum Likelihood Estimation for Training Deep Networks