On complementing end-to-end human behavior predictors with planning
원문 초록
High capacity end-to-end approaches for human motion (behavior) prediction have the ability to represent subtle nuances in human behavior, but struggle with robustness to out of distribution inputs and tail events. Planning-based prediction, on the other hand, can reliably output decent-but-not-great predictions: it is much more stable in the face of distribution shift (as we verify in this work), but it has high inductive bias, missing important aspects that drive human decisions, and ignoring cognitive biases that make human behavior suboptimal. In this work, we analyze one family of approaches that strive to get the best of both worlds: use the end-to-end predictor on common cases, but do not rely on it for tail events / out-of-distribution inputs -- switch to the planning-based predictor there. We contribute an analysis of different approaches for detecting when to make this switch, using an autonomous driving domain. We find that promising approaches based on ensembling or generative modeling of the training distribution might not be reliable, but that there very simple methods which can perform surprisingly well -- including training a classifier to pick up on tell-tale issues in predicted trajectories.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Temporal Continual Learning with Prior Compensation for Human Motion Prediction
Stochastic Human Motion Prediction with Memory of Action Transition and Action Characteristic
Multi-Resolution Haar Network: Enhancing human motion prediction via Haar transform
FocalAD: Local Motion Planning for End-to-End Autonomous Driving
TrajFlow: Multi-modal Motion Prediction via Flow Matching
Rodrigues Network for Learning Robot Actions
토론
이 논문과 연결된 포럼 글이 아직 없습니다.