motion planning
Reinforcement Learning (RL) has shown excellent performance in solving decision-making and control problems of autonomous driving, which is increasingly applied in diverse driving scenarios. However, driving is a multi-attribute problem, leading to challenges in achieving multi-objective compatibility for current RL methods, especially in both policy execution and policy iteration. We propose a Multi-objective Ensemble-Critic reinforcement learning method with Hybrid Parametrized Action for multi-objective compatible autonomous driving.
- Categories:
75 Views
This dataset is based on the ACFR Five Roundabouts Dataset. The original tracking data of over 23,000 traffic vehicles have been processed with an optimization-based filtering method to combat measurement noise and errors. Smooth velocity and acceleration signals are reconstructed. The processed recordings have then undergone a selection process using DBSCAN to remove the erroneous samples. The remaining samples contained in this dataset are considered representative of how average human drivers approach a roundabout scenario in daily driving.
- Categories:
452 Views