Learning Occlusion-aware Decision-making from Agent Interaction via Active Perception

TL;DR

This paper introduces Pad-AI, a novel reinforcement learning framework that enhances occlusion-aware decision-making in autonomous driving by integrating active perception, prediction, and risk-awareness for better exploration and generalization.

Contribution

Pad-AI is a self-reinforcing framework that uses vectorized environment representation and semantic motion primitives to improve occlusion-aware decision-making in dynamic and static scenarios.

Findings

Outperforms strong baselines in occlusion-rich scenarios

Demonstrates efficient and general perception-aware exploration

Provides risk-aware learning with security guarantees

Abstract

Occlusion-aware decision-making is essential in autonomous driving due to the high uncertainty of various occlusions. Recent occlusion-aware decision-making methods encounter issues such as high computational complexity, scenario scalability challenges, or reliance on limited expert data. Benefiting from automatically generating data by exploration randomization, we uncover that reinforcement learning (RL) may show promise in occlusion-aware decision-making. However, previous occlusion-aware RL faces challenges in expanding to various dynamic and static occlusion scenarios, low learning efficiency, and lack of predictive ability. To address these issues, we introduce Pad-AI, a self-reinforcing framework to learn occlusion-aware decision-making through active perception. Pad-AI utilizes vectorized representation to represent occluded environments efficiently and learns over the semantic motion primitives to focus on high-level active perception exploration. Furthermore, Pad-AI integrates prediction and RL within a unified framework to provide risk-aware learning and security guarantees. Our framework was tested in challenging scenarios under both dynamic and static occlusions and demonstrated efficient and general perception-aware exploration performance to other strong baselines in closed-loop evaluations.