Dynamic Car-Following Model With Jerk Suppression for Highway Autonomous Driving by Ke Liu, Jing Ma, Edmund M.-K. Lai

“…This study proposes a dynamic safe car-following strategy that is based on dynamic adjustment of headway time with jerk suppression. Reinforcement learning models trained with this strategy result in enhanced safety and driving comfort, validated using real driving data from the Next Generation Simulation (NGSIM) I-80 and HighD datasets. …”
Get full text

A New Early Rumor Detection Model Based on BiGRU Neural Network by Xiangning Chen, Caiyun Wang, Dong Li, Xuemei Sun

“…Specifically, in this model, the input data is firstly refined through account filtering and data standardization, then the BiGRU is used to consider the context relationship, and a reinforcement learning algorithm is applied to detection. …”
Get full text

Intelligent scheduling mechanism of time-sensitive network modal in polymorphic network by Sijin YANG, Lei ZHUANG, Yu SONG, Jiaxing WANG, Xinyu YANG

“…For the problems of uncertain forwarding scheduling and long solving time of time-sensitive network modal in polymorphic network, a joint routing and scheduling mechanism of time-sensitive network modal based on CSQF was proposed.Considering the requirement of bounded delay, network state and different routing mechanisms, a hybrid resource scheduling problem of joint cache queue and routing was formulated to optimize the resource usage of the entire network.Then, the traffic characteristics and cache queue utilization was used to predict the cache utilization of the next cycle, which was based on deep reinforcement learning.In addition, by using multi-queue CSQF forwarding scheduling mechanism and explicit routing algorithm based on cache utilization, an iterative scheduling algorithm was proposed to achieve deterministic forwarding and resource allocation.Simulation results show that the mechanism can effectively adjust the transmission scheduling of deterministic applications according to the resource usage of the network, and has better schedulability compared with other off-line scheduling mechanisms.…”
Get full text

A Strong Maneuvering Target-Tracking Filtering Based on Intelligent Algorithm by Jing Li, Xinru Liang, Shengzhi Yuan, Haiyan Li, Changsheng Gao

“…The algorithm can map the selection of the model set to the selection of the action label and realize the purpose of a deep reinforcement-learning agent to replace the model switching in the traditional VSMM algorithm by reasonably designing a reward function, state space, and network structure. …”
Get full text

A computing allocation strategy for Internet of things’ resources based on edge computing by Zengrong Zhang

“…In order to meet the demand for efficient computing services in big data scenarios, a cloud edge collaborative computing allocation strategy based on deep reinforcement learning by combining the powerful computing capabilities of cloud is proposed. …”
Get full text

Joint optimization algorithm for 6G network task offloading and fine-grained slice resource scheduling by WANG Ye, WANG Yifei, CHEN Kang, ZHU Xiaorong, TONG En, XU Yufei

“…Then the A3C reinforcement learning algorithm of asynchronous training was used to solve it. …”
Get full text

Coordinated air-ground scheduling for UAV-assisted cell-free massive MIMO by Danhao DENG, Chaowei WANG, Fan JIANG, Weidong WANG

“…Cell-free massive multiple-input multiple-output (MIMO) technology uses a large number of access points (AP) to provide efficient communication services for terrestrial users.However, massive APs and users increase the demand for channel state information detection, especially for users with high moving speeds.In order to reduce the consumption of pilot resources, a dual system architecture that integrates unmanned aerial vehicle-assisted (UAV-assisted) communications and cell-free massive MIMO communications was proposed.The architecture was able to predict the movement trajectories of high-speed users and use UAVs to provide them with reliable communications.A UAV trajectory design and user scheduling scheme based on deep reinforcement learning (DRL) was further proposed, which maximized the system sum rate under the premise of satisfying various constraints.Simulation results demonstrate that the proposed scheme is able to predict user trajectories and improve system sum rate compared with existing schemes.…”
Get full text

Convolutional neural network (CNN) configuration using a learning automaton model for neonatal brain image segmentation. by Iran Sarafraz, Hamed Agahi, Azar Mahmoodzadeh

“…., size, length, and width of the filter in each layer along with the type of pooling functions with a reinforcement learning approach and an LA model are determined. …”
Get full text

Trajectory planning of UUV-assisted UWOC systems based on DQN by Jiawei HU, Xiaoqian LIU, Xinke TANG, Yuhan DONG

“…As a key submarine-based communication platform, unmanned underwater vehicle (UUV) can facilitate underwater wireless optical communication (UWOC).However, fluctuating characteristics of water body, different water qualities, multi-user access present challenges to UUV-assisted UWOC systems, which could be alleviated by an appropriate path planning to maximize the system and each user performance.Deep reinforcement learning (DRL) technology was applied in the path planning of autonomous vehicles, a trajectory planning scheme for UUV-assisted UWOC systems was proposed.The UUV automatically decides the navigation direction through deep Q-network (DQN) method, thereby improving the communication capacity of the system and each user.The impact of distinct water qualities on the capacity enhancement was also investigated.Simulation results suggest that the outputted strategy of DQN can improve the link capacity of the system and each user.This capacity improvement in clear seawater is better than that in pure seawater but lower than that in coastal water.…”
Get full text

Joint optimization method of intelligent service arrangement and computing-networking resource allocation for MEC by Yun LI, Qian GAO, Zhixiu YAO, Shichao XIA, Jishen LIANG

“…., computing power, storage, and bandwidth, with the objective of minimizing task processing latency, the joint optimization of service caching and computing-networking resource allocation was abstracted as a partially observable Markov decision process.Considering the temporal dependency of service request and its coupling relationship with service caching, a long short-term memory network was introduced to capture time-related network state information.Then, based on recurrent multi-agent deep reinforcement learning, a distributed service arrangement and resource allocation algorithm was proposed to autonomously decide service caching and computing-networking resource allocation strategies.Simulation results demonstrate that significant performance improvements in terms of cache hit rate and task processing latency achieved by the proposed algorithm.…”
Get full text

Data-Driven Robust Optimization of the Vehicle Routing Problem with Uncertain Customers by Jingling Zhang, Yusu Sun, Qinbing Feng, Yanwei Zhao, Zheng Wang

“…The Q-learning algorithm in reinforcement learning is introduced into the high-level selection strategy using the hyper-heuristic algorithm, and a hyper-heuristic algorithm based on the Q-learning algorithm is designed to solve the problem. …”
Get full text

Bending obstacles when moving a mobile robot by A. V. Sidorenko, N. A. Saladukha

“…The developed software includes the Mobile Robotics Simulation Toolbox, Reinforcement Learning Toolbox, and the Gazebo visualization package for environment simulation. …”
Get full text

Environmental regulation, market power and low-carbon development of China's coal power industry chain —Based on both strategy and return perspectives by Jiaming Gao, Li Zhang

“…Furthermore, a reinforcement learning model has been developed using the payoff matrix. …”
Get full text

Prioritized Experience Replay–Based Path Planning Algorithm for Multiple UAVs by Chongde Ren, Jinchao Chen, Chenglie Du

“…In this paper, we study the path planning problem for multiple UAVs and propose a reinforcement learning algorithm: PERDE-MADDPG based on prioritized experience replay (PER) and delayed update skills. …”
Get full text

Context-aware learning-based access control method for power IoT by Zhenyu ZHOU, Zehan JIA, Haijun LIAO, Xiongwen ZHAO, Lei ZHANG

“…In view of the problems of severe access conflicts, high queue backlog, and low energy efficiency in the massive terminal access scenario of the power Internet of things (power IoT) in 6G era, a context-aware learning-based access control (CLAC) algorithm was proposed.The proposed algorithm was based on reinforcement learning and fast uplink grant technology, considering active state and dormant state of terminals, and the optimization objective was to maximize the total network energy efficiency under the long-term constraint of terminal access service quality requirements.Lyapunov optimization was used to decouple the long-term optimization objective and constraint, and the long-term optimization problem was transformed into a series of single time-slot independent deterministic sub-problems, which could be solved by the terminal state-aware upper confidence bound algorithm.The simulation results show that CLAC can improve the network energy efficiency while meeting the terminal access service quality requirements.Compared with the traditional fast uplink grant, CLAC can improve the average energy efficiency by 48.11%, increase the proportion of terminals meeting access service quality requirements by 54.95%, and reduce the average queue backlog by 83.83%.…”
Get full text

The Ecosystem of AI-Driven Robotics in Pediatric Neurorehabilitation by Tole Sutikno, Lina Handayani

“…We address this review to facilitate an in-depth analysis of the effective integration of advanced technologies, such as artificial emotional intelligence and interactive reinforcement learning, into rehabilitation practices. By critically assessing each element, from the psychological dynamics of patient engagement to the technical intricacies of real-time adaptive learning systems, we can better understand their pivotal roles in enhancing therapeutic efficacy. …”
Get full text

DDPG-based performance optimization algorithm for IRS-assisted simultaneous wireless information and power transfer systems by LUO Liping, PAN Weimin

“…To solve the non-convex optimization problem, a deep deterministic policy gradient (DDPG) algorithm based on deep reinforcement learning was proposed. Simulation results show that the average reward of the DDPG algorithm is related to the learning rate. …”
Get full text

Q-Learning-Driven Butterfly Optimization Algorithm for Green Vehicle Routing Problem Considering Customer Preference by Weiping Meng, Yang He, Yongquan Zhou

“…This paper proposes a Q-learning-driven butterfly optimization algorithm (QLBOA) by integrating the Q-learning mechanism of reinforcement learning into the butterfly optimization algorithm (BOA). …”
Get full text

Research Status on Jumping Function of Quadruped Robots by Hua Congcong, Yang Zihe, Xia Yanting, Yan Jianrong, Gao Guanjian, Luo Fuliang, Ma Wenshuo, Li Zhijie

“…In addition, advanced control methods such as reinforcement learning, neural networks, and genetic algorithms also were explored and applied. …”
Get full text

Research on heterogeneous radio access and resource allocation algorithm in vehicular fog computing by Kai XIONG, Supeng LENG, Ke ZHANG, Hao LIU

“…With the development of intelligent transportation and the constant emergence of new vehicular on-board applications,such as automatic driving,intelligent vehicular interaction and safety driving.It is difficult for an independent vehicle to run a wide variety of applications with a large number of computing needs and time delay needs relying on its own limited computing resources.By distributing computing tasks in devices on the edge of the network,fog computing applies virtualization technology,distributed computing technology and parallel computing technology to enable users to dynamically obtain computing power,storage space and other services on demand.Applying fog computing architecture to Internet of vehicles can effectively alleviate the contradiction between the large computing-low delay demands and limited vehicular resources.By analyzing the channel capacity of vehicle-to-vehicle communication,vehicle-infrastructure communication and vehicle-time-delay tolerant network communication,an optimization model of heterogeneous access to multi-service resources for the Internet of vehicles was established,and various vehicle-to-fog resources were jointly dispatched to realize efficient processing of intelligent transportation applications.The simulation results show that the proposed reinforcement learning algorithm can effectively deal with the resource allocation in the heterogeneous vehicular fog architecture.…”
Get full text