scholarly journals Intelligent Network Traffic Control Based on Deep Reinforcement Learning

2022 ◽  
Vol 16 ◽  
pp. 585-594
Author(s):  
Fei Wu ◽  
Ting Li ◽  
Fucai Luo ◽  
Shulin Wu ◽  
Chuanqi Xiao
Keyword(s):  
Reinforcement Learning ◽  
Flow Control ◽  
Network Traffic ◽  
Traffic Control ◽  
Data Center ◽  
Control Method ◽  
Learning Algorithm ◽  
Traffic Density ◽  
Intelligent Network ◽  
Application Process

This paper studies the problems of load balancing and flow control in data center network, and analyzes several common flow control schemes in data center intelligent network and their existing problems. On this basis, the network traffic control problem is modeled with the goal of deep reinforcement learning strategy optimization, and an intelligent network traffic control method based on deep reinforcement learning is proposed. At the same time, for the flow control order problem in deep reinforcement learning algorithm, a flow scheduling priority algorithm is proposed innovatively. According to the decision output, the corresponding flow control and control are carried out, so as to realize the load balance of the network. Finally, experiments show, the network traffic bandwidth loss rate of the proposed intelligent network traffic control method is low. Under the condition of random 60 traffic density, the average bisection bandwidth obtained by the proposed intelligent network traffic control method is 4.0mbps and the control error rate is 2.25%. The intelligent network traffic control method based on deep reinforcement learning has high practicability in the practical application process, and fully meets the research requirements.

scholarly journals Macroscopic Traffic Flow Control Via State Estimation

2015 ◽  
Vol 15 (5) ◽  
pp. 5-16
Author(s):  
H. Abouaïssa ◽  
H. Majid
Keyword(s):  
Flow Control ◽  
State Estimation ◽  
Traffic Flow ◽  
Traffic Control ◽  
Traffic Density ◽  
Traffic State ◽  
Loop Detectors ◽  
Traffic State Estimation ◽  
Estimation And Control ◽  
And Control

Abstract The studies presented in this paper deal with traffic control in case of missing data and/or when the loop detectors are faulty. We show that the traffic state estimation plays an important role in traffic prediction and control. Two approaches are presented for the estimation of the main traffic variables (traffic density and mean speed). The state constructors obtained are then used for traffic flow control. Several numerical simulations show very promising results for both traffic state estimation and control.

A Novel Non-Supervised Deep-Learning-Based Network Traffic Control Method for Software Defined Wireless Networks

2018 ◽  
Vol 25 (4) ◽  
pp. 74-81 ◽  
Author(s):  
Bomin Mao ◽  
Fengxiao Tang ◽  
Zubair Md. Fadlullah ◽  
Nei Kato ◽  
Osamu Akashi ◽  
...  
Keyword(s):  
Wireless Networks ◽  
Deep Learning ◽  
Network Traffic ◽  
Traffic Control ◽  
Control Method

Optimal Signal Timing of Single Intersection for Traffic Emission Control

2014 ◽  
Vol 587-589 ◽  
pp. 2137-2140
Author(s):  
Xin Li ◽  
Feng Chen
Keyword(s):  
Reinforcement Learning ◽  
Traffic Control ◽  
Learning Algorithm ◽  
Emission Control ◽  
Simulation Software ◽  
Atmospheric Environment ◽  
Traffic Signal Control ◽  
Traffic Emission ◽  
Average Delay ◽  
Control Scheme

Traffic emission is one of the main pollution sources of urban atmospheric environment. Traffic control scheme of intersection has important influence on vehicle emission. Research on low emission traffic signal control scheme has become one of focuses of Intelligent Transportation. Current typical control methods of traffic emission are based on optimizing the average delay and number of stops. However, it is extremely difficult to use mathematical formula to calculate the delay and the number of stops in the presence of initial queue length of intersection. In order to solve this problem, we proposed a traffic emission control algorithm based on reinforcement learning. The simulation experiments were carried out by using the microscopic traffic simulation software. Compared with the Hideki emission control scheme, the experimental results show that the reinforcement learning algorithm is more effective. The average vehicle emissions are reduced by 12.2% for high saturation of the intersection.

State-of-the-Art Deep Learning: Evolving Machine Intelligence Toward Tomorrow’s Intelligent Network Traffic Control Systems

2017 ◽  
Vol 19 (4) ◽  
pp. 2432-2455 ◽  
Author(s):  
Zubair Md. Fadlullah ◽  
Fengxiao Tang ◽  
Bomin Mao ◽  
Nei Kato ◽  
Osamu Akashi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Control Systems ◽  
Network Traffic ◽  
Traffic Control ◽  
State Of The Art ◽  
Machine Intelligence ◽  
Intelligent Network

scholarly journals A Deep Reinforcement Learning Approach to the Optimization of Data Center Task Scheduling

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-12 ◽  
Author(s):  
Haiying Che ◽  
Zixing Bai ◽  
Rong Zuo ◽  
Honglei Li
Keyword(s):  
Reinforcement Learning ◽  
Task Scheduling ◽  
Data Center ◽  
Delay Time ◽  
Performance Metrics ◽  
Heuristic Algorithms ◽  
Learning Algorithm ◽  
Real Data ◽  
Average Delay ◽  
Scheduling Model

With more businesses are running online, the scale of data centers is increasing dramatically. The task-scheduling operation with traditional heuristic algorithms is facing the challenges of uncertainty and complexity of the data center environment. It is urgent to use new technology to optimize the task scheduling to ensure the efficient task execution. This study aimed at building a new scheduling model with deep reinforcement learning algorithm, which integrated the task scheduling with resource-utilization optimization. The proposed scheduling model was trained, tested, and compared with classical scheduling algorithms on real data center datasets in experiments to show the effectiveness and efficiency. The experiment report showed that the proposed algorithm worked better than the compared classical algorithms in the key performance metrics: average delay time of tasks, task distribution in different delay time levels, and task congestion degree.

scholarly journals On Stability of Perturbed Nonlinear Switched Systems with Adaptive Reinforcement Learning

Energies ◽  
2020 ◽  
Vol 13 (19) ◽  
pp. 5069
Author(s):  
Phuong Nam Dao ◽  
Hong Quang Nguyen ◽  
Minh-Duc Ngo ◽  
Seon-Ju Ahn
Keyword(s):  
Optimal Control ◽  
Reinforcement Learning ◽  
Switched Systems ◽  
Tracking Control ◽  
Control Method ◽  
Learning Algorithm ◽  
Hjb Equation ◽  
Control Approach ◽  
Nonlinear Switched Systems ◽  
Hamilton Jacobi Bellman

In this paper, a tracking control approach is developed based on an adaptive reinforcement learning algorithm with a bounded cost function for perturbed nonlinear switched systems, which represent a useful framework for modelling these converters, such as DC–DC converter, multi-level converter, etc. An optimal control method is derived for nominal systems to solve the tracking control problem, which results in solving a Hamilton–Jacobi–Bellman (HJB) equation. It is shown that the optimal controller obtained by solving the HJB equation can stabilize the perturbed nonlinear switched systems. To develop a solution to the translated HJB equation, the proposed neural networks consider the training technique obtaining the minimization of square of Bellman residual error in critic term due to the description of Hamilton function. Theoretical analysis shows that all the closed-loop system signals are uniformly ultimately bounded (UUB) and the proposed controller converges to optimal control law. The simulation results of two situations demonstrate the effectiveness of the proposed controller.

scholarly journals A Disturbance Rejection Control Method Based on Deep Reinforcement Learning for a Biped Robot

2021 ◽  
Vol 11 (4) ◽  
pp. 1587
Author(s):  
Chuzhao Liu ◽  
Junyao Gao ◽  
Dingkui Tian ◽  
Xuefeng Zhang ◽  
Huaxin Liu ◽  
...  
Keyword(s):  
Reinforcement Learning ◽  
Disturbance Rejection ◽  
Humanoid Robot ◽  
Control Method ◽  
Learning Algorithm ◽  
Biped Robot ◽  
Forward Direction ◽  
Disturbance Rejection Control ◽  
Table Model ◽  
External Forces

The disturbance rejection performance of a biped robot when walking has long been a focus of roboticists in their attempts to improve robots. There are many traditional stabilizing control methods, such as modifying foot placements and the target zero moment point (ZMP), e.g., in model ZMP control. The disturbance rejection control method in the forward direction of the biped robot is an important technology, whether it comes from the inertia generated by walking or from external forces. The first step in solving the instability of the humanoid robot is to add the ability to dynamically adjust posture when the robot is standing still. The control method based on the model ZMP control is among the main methods of disturbance rejection for biped robots. We use the state-of-the-art deep-reinforcement-learning algorithm combined with model ZMP control in simulating the balance experiment of the cart–table model and the disturbance rejection experiment of the ASIMO humanoid robot standing still. Results show that our proposed method effectively reduces the probability of falling when the biped robot is subjected to an external force in the x-direction.

Sign in / Sign up

Export Citation Format

Share Document