In the past, most robots used rigid structures. With the development of intelligent materials, more and more soft materials began to be used in the manufacture of robots. Magnetic-driven soft robot is one of its impor...
详细信息
In the past, most robots used rigid structures. With the development of intelligent materials, more and more soft materials began to be used in the manufacture of robots. Magnetic-driven soft robot is one of its important branches. Because of the flexibility of magnetoelastic composite robots and the harmlessness of magnetic field to human body, a magnetically-controlled soft robot has broad application prospects in medical fields such as disease diagnosis and treatment in human body. A small lamellate magnetic soft robot is designed in this paper with the advantages of small volume, light weight, high deformation, fast and simple manufacture. And it can be driven without wire under the control of the magnetic field formed by the three-dimensional Helmholtz coil. The robot can walk and crawl on the surface of magnetically-controlled. The motion characteristics of the magnetic soft robot such as crawling step length and bending height are analyzed. The data are collected through experiments to analyze the relationship between the control signal and the motion characteristics.
Reliable fault diagnosis is of practical significance for process safety in modern industrial facilities. This article proposes a data-driven fault diagnosis method based on multi-feature fusion and broad learning. Th...
详细信息
This paper proposes a distributed optimization algorithm based on alternating direction method of multipliers (ADMM) for the distributed optimization problem of multi-agent systems, called ADMM with adaptive penalty t...
详细信息
In this paper, an exponential synchronization design method for chaotic Lur'e systems is developed. Firstly, a sampled-data based proportional-integral (PI) type controller is used in the slave system to achieve t...
详细信息
In this paper, an exponential synchronization design method for chaotic Lur'e systems is developed. Firstly, a sampled-data based proportional-integral (PI) type controller is used in the slave system to achieve the synchronization. Next, the synchronization problem of the chaotic Lur'e system is transformed into the stabilization problem of the synchronization error system. Then, by introducing virtual vectors for reconstruction, a new augmented error system is obtained. As a result, the synchronization design based on sampled-data PI type controller can be achieved using the existing sampling system method. And based on this, a criterion for exponential synchronization design is obtained using the Lyapunov functional method. Finally, the effectiveness of the developed method is demonstrated by numerical example.
Aiming at the ore blending problem in the open-pit mining scenario, the optimization objective is to minimize the grade variance of the sliding window of the ore flow, and a cooperative shovel scheduling model for rea...
详细信息
ISBN:
(数字)9798350368604
ISBN:
(纸本)9798350368611
Aiming at the ore blending problem in the open-pit mining scenario, the optimization objective is to minimize the grade variance of the sliding window of the ore flow, and a cooperative shovel scheduling model for real-time ore blending is established. A reinforcement learning-enhanced dynamic multi- objective evolutionary algorithm is proposed, where a Q-learning operator selection mechanism is introduced to reuse the information from the previous environment for tracking the optimal solution in the dynamic environment. Experimental results show that the proposed method can effectively control the grade fluctuation of the ore flow and dynamically respond to external events, and can find a better solution in real time compared with the traditional operator selection algorithm.
作者：
Zexi WangWei XueKehui ChenShaopeng MaSchool of Automation
China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan China
In recent years, deep neural network has continuously improved the performance of remote sensing image classification. Though The deep neural network model is powerful, it is difficult to deploy on resource-constraine...
详细信息
In recent years, deep neural network has continuously improved the performance of remote sensing image classification. Though The deep neural network model is powerful, it is difficult to deploy on resource-constrained hardware platforms such as mobile terminal devices and embedded devices due to the large number of network parameters. In this paper, a novel lightweight network (LW-Net) model and a network pruning method are used for remote sensing image classification. This LW-Net model adopts a net block unit to obtain more characteristic graphs with less computational complexity. The proposed network pruning method uses the sparsity regularization on the influence factor in BN layer to automatically identified and pruned unimportant channels to make the model structure simpler. Experimental results demonstrate compared with traditional deep neural networks, the proposed model with the network pruning method can greatly reduce the computational complexity and model parameters with a little accuracy loss.
With the development of radar technology, frequency modulated continuous wave (FMCW) radar has been used for non-contact vital signs detection. In order to suppress the environmental noise and interference of breathin...
详细信息
With the development of radar technology, frequency modulated continuous wave (FMCW) radar has been used for non-contact vital signs detection. In order to suppress the environmental noise and interference of breathing harmonics on heartbeat signal, this paper proposes a new vital sign detection method based on scale-space representation (SSR) and empirical wavelet transform (EWT) for the FMCW radar. First, a low-pass filter is set for the high-frequency noise elimination outside the frequency range of vital sign signals. Second, SSR is used to adaptively segment the spectrum of signals to obtain the initial frequency boundaries, and the kurtosis of the spectrum is applied to segment the spectrum again and obtain the final frequency boundaries to further eliminate noise. Third, the empirical wavelet filter bank is constructed by using the determined boundaries and EWT is used to decompose the signals to several components. Finally, according to the frequency ranges of components and the correlation between the components and vital sign signals, the components are selected to reconstruct the breathing and heartbeat signals. The experimental results show that the proposed method achieves a better detection performance than the classical EWT and empirical mode decomposition (EMD).
An adaptive backstepping dynamic surface sliding mode controller based on nonlinear disturbance observer (NDOABSMC) is designed to track zigzag motion of underwater glider (UG). Firstly, a nonlinear disturbance observ...
An adaptive backstepping dynamic surface sliding mode controller based on nonlinear disturbance observer (NDOABSMC) is designed to track zigzag motion of underwater glider (UG). Firstly, a nonlinear disturbance observer is devised to observe ocean current. Then, the backstepping sliding mode control is used to devise motion attitude controller of UG to ensure that UG can track the target trajectory quickly. Additionally, the stability of UG's controlsystem is analyzed. In the end, the proposed control strategy is compared with other approaches.
With the development of unmanned surface vessel in the world today, obstacle avoidance using environmental information is the basis to ensure its high maneuvering performance and safety. However, directly using standa...
详细信息
With the development of unmanned surface vessel in the world today, obstacle avoidance using environmental information is the basis to ensure its high maneuvering performance and safety. However, directly using standard algorithms will lead to missing and wrong identification severely for characteristics of marine obstacles. This paper adds a multi-scale feature extraction layer of dilation convolution and group convolution to Faster Region based Convolutional Neural Network (Faster-RCNN), the baseline model, and changes the classification algorithm to improve its robustness and accuracy. Soft-Non Maximum Suppression (Soft-NMS) is used to enhance the prediction effects further. After improvements, the mean average precision value increases by 3.35%, and the final loss value decreases by 0.20. Given the phenomenon of missing and misidentification in the prediction by the baseline model, the results of our new model show outstanding performance.
In this paper, the effect of the initial state of the drilling system on trajectory tracking is taken into account in the directional drilling trajectory tracking control. Firstly, the trajectory model describing the ...
详细信息
ISBN:
(数字)9789887581598
ISBN:
(纸本)9798331540845
In this paper, the effect of the initial state of the drilling system on trajectory tracking is taken into account in the directional drilling trajectory tracking control. Firstly, the trajectory model describing the two-dimensional drilling trajectory of the rotary steering system (RSS) is introduced. To improve the accuracy of trajectory tracking control and reduce the trajectory deviation, a fixed-time sliding mode control (FTSMC) method is introduced. Then, the stability of the closed-loop system is analyzed, and the range of controller parameters satisfying the stability condition is obtained. Finally, the effectiveness of the designed control strategy is verified by simulation experiments.
暂无评论