… behavior planning structure based on reinforcementlearning (RL) which is capable of performing autonomous vehiclebehavior planning with a hierarchical … of the hierarchical structure […
Y Lu, X Xu, X Zhang, L Qian, X Zhou - IEEE Access, 2020 - ieeexplore.ieee.org
… vehicles. Aiming at the above problems, this article proposes a hierarchicalreinforcement learningapproach … The proposed approach consists of two layers. At the higher layer, a kernel-…
H Zhang, J Peng, H Dong, H Tan, F Ding - Applied Energy, 2023 - Elsevier
… Recent advances in the ecological car following issue of HEV … a hierarchicalreinforcement learning based ACC-EMS strategy is proposed with a hierarchical policy and non-hierarchical …
Y Chen, C Dong, P Palanisamy… - Proceedings of the …, 2019 - openaccess.thecvf.com
… racing cars. In this work, A deep reinforcementlearning (DRL) with a novel hierarchical structure … Hierarchical actions for multi-scale tasks have been actively studied in DRL [12, 27]. [8] …
… Supervised learningmethods such as Imitation Learning lack generalization and safety … a RobustHierarchicalReinforcementLearning (HRL) framework for learning autonomous driving …
J Chen, Z Wang, M Tomizuka - 2018 IEEE intelligent vehicles …, 2018 - ieeexplore.ieee.org
… acteristics of hierarchicalreinforcementlearning to separate these different behaviors into different policies. Then we train an additional master policy that can choose which behavior to …
… depends not only on historical information, but also on future plans of other vehicles that … probabilistic prediction approach based on hierarchical inverse reinforcementlearning (IRL)…
Z Gu, L Gao, H Ma, SE Li, S Zheng… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
… In sum, we propose the state-based safety enhancement method for autonomous vehicles via the direct hierarchicalreinforcementlearning (Safari), and verify it in the highway scenario …
… a general behavior planner for autonomous vehicles based on reinforcementlearning and … By combining the power of lowlevel safe controllers with a high-level reinforcementlearning …