Problem
Research questions and friction points this paper is trying to address.
behavior planning
automated driving
safety assurance
explainability
real-world deployment
Learning-Based Behavior Planning for Automated Driving: Real-World Integration and Deployment
🤖 AI Summary
This work addresses the limitations of learning-based behavioral planning in autonomous driving—particularly poor interpretability and challenges in ensuring safety—by proposing a hybrid planning architecture. The approach integrates the perceptual strengths of deep neural networks in complex traffic scenarios with an optimization-based supervisory layer that validates high-level behavior proposals and enforces feasibility and safety constraints. By synergistically combining data-driven flexibility with classical, deterministic planning principles, the method significantly enhances verifiability and safety without sacrificing adaptability. The system is evaluated through open-loop testing on real-world urban driving data and demonstrates stable closed-loop performance, having been successfully deployed on the research vehicle Karl, thereby confirming its practicality and reliability.
📝 Abstract
Recent research in machine and deep learning has shown the potential of learningbased motion planning approaches to improve the driving behavior of automated vehicles, especially in complex environments. However, their complex nature and lack of transparency can hinder explainability and trustworthiness and complicate safety assurance. Motivated by these challenges, we propose a hybrid planning architecture that combines the advantages of machine learning with the verifiability and the determinism of classical approaches. Specifically, we developed a deep neural network to interpret complex traffic scenes and propose driving behavior, while an optimization-based supervision layer validates this proposal and enforces explicit drivability and safety constraints. We evaluate the learned planner's driving behavior in open-loop studies on real-world urban data, discuss system integration aspects for stable closed-loop operation, and report results from real-world deployment on our research vehicle karl..
Innovation
Methods, ideas, or system contributions that make the work stand out.
hybrid planning architecture
deep neural network
optimization-based supervision
safety constraints
real-world deployment
🔎 Similar Papers
No similar papers found.
