This paper focuses on the security consensus problem of multi-agent systems under cyber-attack. Two types of cyber-attack are considered in our research, this situation is more common in real applications. These two t...
详细信息
ISBN:
(纸本)9798350377859;9798350377842
This paper focuses on the security consensus problem of multi-agent systems under cyber-attack. Two types of cyber-attack are considered in our research, this situation is more common in real applications. These two types of network attacks (deception and DoS attacks) may occur alternately, resulting in the infeasibility to obtain actual output measurements. To reach the system security, a novel iterative learningcontrol strategy is designed to analyze the issue. Then, sufficient conditions are proposed from the view of the trajectory tracking, and our achieved convergence conditions are relatively simpler in comparison to the existing literature. Furthermore, the theoretical result in this paper is an extension of existing research. Finally, a numerical simulation is presented for illustration.
Addressing the need for minimal overshoot in the position control process of a hydraulic cylinder, a model-free predictive control algorithm based on limit learning machine without the need of a specific model of the ...
详细信息
The problem of iterative learningcontrol for hyperbolic distributed parameter systems with faults under sensor/actuator networks is studied. Many results have been achieved in previous studies on iterative learning c...
详细信息
Time-varying delay, as one of the most prominent issues existing in teleoperation system, throws a critical threat on the teleoperation system stability. To address this issue, this paper proposes a novel synchronizat...
详细信息
ISBN:
(纸本)9798350321050
Time-varying delay, as one of the most prominent issues existing in teleoperation system, throws a critical threat on the teleoperation system stability. To address this issue, this paper proposes a novel synchronization control for nonlinear uncertain bilateral teleoperation systems in the presence of time-varying delay. Firstly, a filter error variable is designed to avoid using both local and remote acceleration signals. Then, a set of first-order low-pass filter operations are employed to construct an unknown system dynamics estimator (USDE) to handle the system uncertain dynamics, which revolves the Coriolis/gravity dynamics, external disturbances and remote velocity. With the suggested filtered error and USDE, a feedback controller is developed, which can not only improve the system synchronization but also has the capability of accommodating the influence of time-varying delay. Rigorously theoretical analysis is conducted by choosing the Lyapunov-Razumikhin candidate function to prove the stability of the closed-loop system. Finally, simulation results are provided to illustrate the effectiveness of the proposed method.
As the most fundamental type of matrix equation, the Sylvester equation has been widely applied in control theory, signal processing, and many other scientific fields in recent years. The traditional method for solvin...
详细信息
ISBN:
(纸本)9798350321050
As the most fundamental type of matrix equation, the Sylvester equation has been widely applied in control theory, signal processing, and many other scientific fields in recent years. The traditional method for solving the Sylvester equation is to transform it into a linear algebraic equation (LAE). However, this method will lead to an increase in the dimension of the coefficient matrix, which makes it difficult to solve the LAE. To alleviate the above problem, a distributed algorithm for solving the Sylvester equation is presented in this paper. Firstly, we obtain a LAE equivalent to the Sylvester equation by utilizing vectorization operation and Kronecker product. Then, a group of agents in the multi-agent system is considered to implement the distributed solution for LAE, where each agent only solves its local task by constantly exchanging information with its neighbors. By constructing the iterative learningcontrol system, a discrete linear system about the tracking error of the agent is obtained. Based on the average neighbor information and the feedback control design, an updating rule for each agent iteratively updating its state is obtained. It is shown that all agents converge to the vectorization solution of the Sylvester equation when the communication topology between agents is undirected complete graph. Finally, a simulation example is provided to demonstrate the effectiveness of the proposed distributed algorithm.
Deep reinforcement learning algorithms have made great progress in the field of control with the help of many high-efficiency simulation environments. However, due to the difference in state distribution and dynamics,...
详细信息
ISBN:
(纸本)9798350321050
Deep reinforcement learning algorithms have made great progress in the field of control with the help of many high-efficiency simulation environments. However, due to the difference in state distribution and dynamics, these algorithms trained in the simulation cannot be effectively applied to the real world. The ability to reduce the impact of the Sim2Real gap is critical for transferring policy from the simulation to the real world. Although there are many methods for studying the Sim2Real problem, it is difficult to evaluate the performance of different algorithms due to the different evaluation platforms and evaluation metrics. In this paper, we construct a uniform robot navigation scenario, and revisit the ability of the popular domain randomization methods to transfer the policies from the simulation to the real world under the dynamics gaps. With the analysis of the performence in the simulation environment and the real world, we provide some recommendations of the domain randomization methods and hope to make these methods more efficient to use.
In many basic oxygen furnace (BOF) steelmaking processes, if the furnace endpoint carbon can be monitored in real time, it is a breakthrough for BOF steelmaking intelligence. This paper presents a deep learning model ...
详细信息
In most adaptive iterative learningcontrol (AILC) researches, there are many achievements in one-dimensional (1-D) dynamic systems, but few in two-dimensional dynamic systems. Considering the unknown control directio...
详细信息
In the scenario where cameras are set up on the tower to capture the whole image of heliostat field, traditional image processing algorithms often fail to correctly identify the heliostats, due to issues such as over ...
详细信息
As a vulnerability discovery technique, fuzzing has been widely used in the field of software test in the past years. Traditional fuzzing has several drawbacks, including poor efficiency, low code coverage, and a high...
详细信息
ISBN:
(纸本)9798350321050
As a vulnerability discovery technique, fuzzing has been widely used in the field of software test in the past years. Traditional fuzzing has several drawbacks, including poor efficiency, low code coverage, and a high dependence on expert experience. By introducing the deep reinforcement learning technique, one can train the mutator of the fuzzer to move in a desired direction, such as maximizing code coverage or finding more code paths. This paper proposes a reinforcement learningbased fuzzing method to enhance the code coverage and explore potential code vulnerabilities. First, the concept of the input field is introduced to the seed file, reducing invalid operations by marking whether each byte of the seed file is a valid byte. Then, we optimize mutation by modeling the grey-box fuzzing as a reinforcement learning problem and training mutator's behavior on test cases. By observing the rewards caused by mutating with a specific set of actions performed on an initial program input, the fuzzing agent learns a policy that can next generate new higher-reward inputs. Finally, experimental results show that the proposed deep reinforcement learning-based fuzzing method outperforms the baseline random fuzzing algorithms.
暂无评论