Y Zhang, X Liang, D Li, SS Ge, B Gao… - IEEE Transactions …, 2023 - ieeexplore.ieee.org
… invariably requires that the full-state variables are constrained within the safety region even … an adaptive safereinforcement learning (RL) algorithm that invokes innovative safety-related …
… Abstract—The dependence of autonomousvehicles (AVs) on … and a novel adversarial deep reinforcement learning (RL) algorithm … the inter-vehicle optimal safe spacing and potentially …
LC Das, M Won - International Conference on Machine …, 2021 - proceedings.mlr.press
… safety, and driving comfort through dynamic adaptation of the inter-vehicle gap based on deep reinforcement … and driving safety/comfort by effectively controlling the driving safety model …
… for autonomousvehicles confirm we can ensure safety while … that the policy cannot be guaranteed to be safe, and the back-… Safety is the most basic requirement for autonomousvehicles…
… that the safe decision making is the … autonomous navigation since at any moment the safe decision boundary can be determined by means of kinematic information from nearby vehicles …
Y Hu, J Fu, G Wen - IEEE Transactions on Intelligent Vehicles, 2023 - ieeexplore.ieee.org
… In this paper, we propose a novel safe model-based RL … trajectory tracking problem of uncertain autonomousvehicles (AVs). … safe modelbased RL algorithm is proposed, where the safe …
S Mo, X Pei, C Wu - IEEE Transactions on Intelligent …, 2021 - ieeexplore.ieee.org
… In this paper, we propose a safereinforcement learning framework combined with … safe reinforcement learning approach in simulation scenario built in SUMO. Autonomousvehicle …
HD Nguyen, K Han - International Journal of Control, Automation and …, 2023 - Springer
… : Safe decision-making strategy of autonomousvehicles (AVs… This study develops a safe reinforcement learning (safe-RL)-… the proposed safe-RL, where an upper layer executes a safe …
Y Zhang, X Liang, D Li, SS Ge, B Gao… - … on Neural Networks …, 2022 - ieeexplore.ieee.org
… practically challenging for the wide deployment of autonomousvehicles. Safety-critical systems in general, require safe performance even during the reinforcement learning (RL) period. …