Filters








2,727 Hits in 12.7 sec

A Safe Hierarchical Planning Framework for Complex Driving Scenarios based on Reinforcement Learning [article]

Jinning Li, Liting Sun, Jianyu Chen, Masayoshi Tomizuka, Wei Zhan
2021 arXiv   pre-print
Safety is guaranteed by the low-level optimization/sampling-based controllers, while the high-level reinforcement learning algorithm makes H-CtRL an adaptive and efficient behavior planner.  ...  To address this challenge, we propose a hierarchical behavior planning framework with a set of low-level safe controllers and a high-level reinforcement learning algorithm (H-CtRL) as a coordinator for  ...  Authors of [22] developed SAVED which augmented the safety of model-based reinforcement learning with Value estimation from demonstrations.  ... 
arXiv:2101.06778v2 fatcat:jiaiy7gukfhcdeo7cgfjuysi2y

A Review of the Self-Adaptive Traffic Signal Control System Based on Future Traffic Environment

Yizhe Wang, Xiaoguang Yang, Hailun Liang, Yangdong Liu
2018 Journal of Advanced Transportation  
real-time characteristic, accuracy, and self-learning and therefore will be an important research focus of control method in future due to the property of "model-free" and "self-learning" that well accommodates  ...  Finally, the article concluded that signal control based on multiagent reinforcement learning is a kind of closed-loop feedback adaptive control method, which outperforms many counterparts in terms of  ...  Reinforcement Learning, as a typical "model-free, self-learning" iterative data-driven method, is applicable of regional traffic control based on multiagent reinforcement learning [67] .  ... 
doi:10.1155/2018/1096123 fatcat:ynkoxgm6lnflrpliafo6xf6opy

Carl-Lead: Lidar-based End-to-End Autonomous Driving with Contrastive Deep Reinforcement Learning [article]

Peide Cai, Sukai Wang, Hengli Wang, Ming Liu
2021 arXiv   pre-print
Recently, deep reinforcement learning (DRL) has shown promising results in urban driving scenarios.  ...  Autonomous driving in urban crowds at unregulated intersections is challenging, where dynamic occlusions and uncertain behaviors of other vehicles should be carefully considered.  ...  In this work, we present Carl-Lead, a Lidar-based end-to-end autonomous driving method trained with Contrastive deep reinforcement learning.  ... 
arXiv:2109.08473v1 fatcat:n3cgfodga5ewddt32xvda6oo64

Centralized Conflict-free Cooperation for Connected and Automated Vehicles at Intersections by Proximal Policy Optimization [article]

Yang Guan, Yangang Ren, Shengbo Eben Li, Qi Sun, Laiquan Luo, Koji Taguchi, Keqiang Li
2019 arXiv   pre-print
In this paper, we propose a centralized conflict-free cooperation method for multiple connected vehicles at unsignalized intersection using reinforcement learning (RL) to address computation burden naturally  ...  We firstly incorporate a prior model into proximal policy optimization (PPO) algorithm to accelerate learning process.  ...  Model-based RL Recent model-free reinforcement learning algorithms have proposed incorporating given or learned dynamics models as a source of additional data with the intention of reducing sample complexity  ... 
arXiv:1912.08410v1 fatcat:74cnjirgrvazljnckdelnmeubm

DQ-GAT: Towards Safe and Efficient Autonomous Driving with Deep Q-Learning and Graph Attention Networks [article]

Peide Cai, Hengli Wang, Yuxiang Sun, Ming Liu
2022 arXiv   pre-print
skills with them, such as yielding, merging and taking turns, to achieve both safe and efficient driving in various settings.  ...  However, they are either implemented with supervised imitation learning (IL), which suffers from dataset bias and distribution mismatch issues, or are trained with deep reinforcement learning (DRL) but  ...  /H-REIL model exhibits low-level interactive driving skills, and tends to collide with other vehicles, leading to a very low success rate (1∼2%).  ... 
arXiv:2108.05030v2 fatcat:fi4bjyfm5zexlnjs4t6hujv7ou

Navigating an Automated Driving Vehicle via the Early Fusion of Multi-Modality

Malik Haris, Adam Glowacz
2022 Sensors  
On the one hand, modular pipelines break down the driving model into submodels, such as perception, maneuver planning and control.  ...  The autonomous vehicle is equipped with a camera and active sensors, such as LiDAR and Radar, for safe navigation.  ...  A second explanation is that urban driving is more difficult than most tasks, especially asynchronous reinforcement learning.  ... 
doi:10.3390/s22041425 pmid:35214327 pmcid:PMC8878300 fatcat:g7ti567yn5h6jbxbazzeoprnt4

Evolutionary reinforcement learning multi-agents system for intelligent traffic light control: new approach and case of study

Mohamed Amine Basmassi, Sidina Boudaakat, Jihane Alami Chentoufi, Lamia Benameur, Ahmed Rebbani, Omar Bouattane
2022 International Journal of Power Electronics and Drive Systems (IJPEDS)  
This paper proposes a self-adapted approach, called evolutionary reinforcement learning multi-agents system (ERL-MA), which combines computational intelligence and machine learning.  ...  The modeling layer uses the intersection modeling using generalized fuzzy graph technique. The decision layer uses two methods: the novel greedy genetic algorithm (NGGA), and the Q-learning.  ...  The objectives of the Q-learning which is a technique of model-free reinforcement learning algorithm are gain experience to make better decisions.  ... 
doi:10.11591/ijece.v12i5.pp5519-5530 fatcat:otlkrkdvcbadxkq27rzrs2yldi

Interpretable End-to-end Urban Autonomous Driving with Latent Deep Reinforcement Learning [article]

Jianyu Chen, Shengbo Eben Li, Masayoshi Tomizuka
2020 arXiv   pre-print
A sequential latent environment model is introduced and learned jointly with the reinforcement learning process.  ...  In this paper, we propose an interpretable deep reinforcement learning method for end-to-end autonomous driving, which is able to handle complex urban scenarios.  ...  learning (RL) [8] - [10] , which learns a policy by self exploration and reinforcement.  ... 
arXiv:2001.08726v3 fatcat:hn3hgpsxvzba3hs6wqrgc7eu7a

Recent Advances in Deep Reinforcement Learning Applications for Solving Partially Observable Markov Decision Processes (POMDP) Problems Part 2—Applications in Transportation, Industries, Communications and Networking and More Topics

Xuanchen Xiang, Simon Foo, Huanyu Zang
2021 Machine Learning and Knowledge Extraction  
Reinforcement Learning (RL) is an approach to simulate the human's natural learning process, whose key is to let the agent learn by interacting with the stochastic environment.  ...  The fact that the agent has limited access to the information of the environment enables AI to be applied efficiently in most fields that require self-learning.  ...  [36] proposed an inverse reinforcement learning (IRL) approach with DQN to extract the rewards for collision-free lane changing. The agent can perform human-like lane changing behavior; Hoel et al.  ... 
doi:10.3390/make3040043 doaj:45bf00de595c44d186fa3d200589c1c5 fatcat:qx4srh7qabgjvd5l6lj6nulhxa

Deep Reinforcement Learning for Autonomous Driving: A Survey [article]

B Ravi Kiran, Ibrahim Sobh, Victor Talpaert, Patrick Mannion, Ahmad A. Al Sallab, Senthil Yogamani, Patrick Pérez
2021 arXiv   pre-print
With the development of deep representation learning, the domain of reinforcement learning (RL) has become a powerful learning framework now capable of learning complex policies in high dimensional environments  ...  This review summarises deep reinforcement learning (DRL) algorithms and provides a taxonomy of automated driving tasks where (D)RL methods have been employed, while addressing key computational challenges  ...  path planning, development of high-level driving policies for complex navigation tasks, scenario-based policy learning for highways, intersections, merges and splits, reward learning with inverse reinforcement  ... 
arXiv:2002.00444v2 fatcat:axj3ohhjwzdrxp6dgpfqvctv2i

Is it Safe to Drive? An Overview of Factors, Challenges, and Datasets for Driveability Assessment in Autonomous Driving [article]

Junyao Guo, Unmesh Kurup, Mohak Shah
2018 arXiv   pre-print
models.  ...  With recent advances in learning algorithms and hardware development, autonomous cars have shown promise when operating in structured environments under good driving conditions.  ...  Many reinforcement learning based approaches have been reported for driving policy adaptation.  ... 
arXiv:1811.11277v1 fatcat:ztrxyydtuveijizfn6a2dmt5ui

TrajGen: Generating Realistic and Diverse Trajectories with Reactive and Feasible Agent Behaviors for Autonomous Driving [article]

Qichao Zhang, Yinfeng Gao, Yikang Zhang, Youtian Guo, Dawei Ding, Yunpeng Wang, Peng Sun, Dongbin Zhao
2022 arXiv   pre-print
In addition, we develop a data-driven simulator I-Sim that can be used to train reinforcement learning models in parallel based on naturalistic driving data.  ...  Realistic and diverse simulation scenarios with reactive and feasible agent behaviors can be used for validation and verification of self-driving system performance without relying on expensive and time-consuming  ...  skills.  ... 
arXiv:2203.16792v1 fatcat:p75sxbq5trajne2bkbuixqczgu

GINK: Graph-based Interaction-aware Kinodynamic Planning via Reinforcement Learning for Autonomous Driving [article]

Se-Wook Yoo, Chan Kim, Jin-Woo Choi, Seong-Woo Kim, Seung-Woo Seo
2022 arXiv   pre-print
In the experiment, we set up a navigation scenario comprising various situations, with CARLA, an urban driving simulator.  ...  Applying reinforcement learning to autonomous driving entails certain challenges, primarily due to massive traffic flows, which change dynamically.  ...  Interaction-aware Motion Planning Recently, several studies [17] - [19] proposed integrating RL and planning to learn driving skills.  ... 
arXiv:2206.01488v2 fatcat:gctqbsizrfajvee5k6clc3z7am

Scanning the Issue

Azim Eskandarian
2021 IEEE transactions on intelligent transportation systems (Print)  
Their method consists of two parts: a simultaneous localization and mapping (SLAM) algorithm based on sparse point clouds (SPC) and a semantic modeling algorithm based on a modified PointNet model.  ...  Deep learning has emerged as a prominent class of techniques for autonomous driving. This article reviews the stateof-the-art techniques using deep neural networks to control autonomous vehicles.  ...  Deep-Reinforcement-Learning-Based Energy Management Strategy for Supercapacitor Energy Storage Systems in Urban Rail Transit Z. Yang, F. Zhu, and F.  ... 
doi:10.1109/tits.2021.3052540 fatcat:wvccn3i32jdaxoov6mibk2tlku

A Reinforcement Learning Approach for Enacting Cautious Behaviours in Autonomous Driving System: Safe Speed Choice in the Interaction With Distracted Pedestrians

Gastone Pietro Rosati Papini, Alice Plebe, Mauro Da Lio, Riccardo Dona
2021 IEEE transactions on intelligent transportation systems (Print)  
Index Terms-Vulnerable road users, neural networks, reinforcement learning, transfer learning, autonomous driving, intelligent speed adaptation.  ...  on learning safe behavior at a highlevel.  ...  Model-free RL, instead, does not require learning a model of the environment and focuses on learning the action-value function, as shown below.  ... 
doi:10.1109/tits.2021.3086397 fatcat:s2tqygxgejhp3mvethfpx6xcqe
« Previous Showing results 1 — 15 out of 2,727 results