Tan sheep is an important industry in Ningxia province, China, and plays a very important role in local agriculture and animal husbandry.However, most of the existing Tan sheep feeding methods relies on human power, which has deficiencies in the efficiency and effectiveness of the operation. To solve the above problems, the navigation system based on RTK and machine vision is designed for Tan sheep intelligent feeding cart, with ultrasonic sensor for collision avoidance. Compared with the existing Tan sheep feeding methods, this research greatly reduces the need for human and material resources, and makes up for the inaccuracy of RTK navigation in bad weather conditions through machine visual navigation. This study has important reference value for similar intelligent navigation feeding cart.
This study presents a new method for super-resolution of depth images by combining fractional calculus and inverse distance interpolation. The method improves edge recognition through fractional differential and solves the problem of missing data in traditional super-resolution algorithms for depth images. Results show that the proposed method outperforms classical interpolation algorithms, with improved PSNR by 3-10dB. The method effectively solves the impact of depth image edge distance on interpolation results, and provides more texture information.
This paper investigates the dynamic electric vehicle routing problem with electric vehicles (DVRP-E). In the DVRP-E, customer demands may emerge stochastically throughout the operational horizon. Moreover, as electric vehicle (EV) batteries remain a key limiting factor for EVs, EVs must return to the depot before batteries run down. An adaptive large neighborhood search algorithm is proposed for its solution, and computational tests are presented to demonstrate its effectiveness and accuracy. The ALNS algorithm has proved itself an effective tool to solve the DVRP-E with up to 1000 customers according to our computational test.
This paper establishes a delayed SIR malicious virus propagation model in mobile wireless sensor networks (MWSNs). Since these nodes are mobile, we introduce diffusion effect into the model. With time delay as the bifurcation parameter, the local stability of the positive equilibrium point and the existence of Hopf bifurcation are analyzed. We find that when the time delay exceeds the critical value, the system will lose stability and Hopf bifurcation will occur. Eventually, the correctness of the main results is illustrated by numerical simulations.
This article investigates the couple-group secure consensus and detection of heterogeneous multi-agent systems (HMASs) with cooperative-competitive relation and deception attacks. According to cooperative-competitive relation, a new couple-group secure consensus protocol is designed. By applying the graph theory, linear algebra theory, probability theory and Getschgorin theorem, several results have been given to ensure the success of the couple-group secure consensus for this system. Based on these obtained results and T-test, a novel detection algorithm has been put forward to determine the security nodes or not. Several examples have been given to prove the effectiveness of these obtained results and the improved secure detection algorithm.
Many existing methods of forecasting the state- of-health(SOH) assume that training and testing data follow the same distribution. The model based on dataset under one working condition may be ineffective for the dataset under another working condition due to the distribution discrepancy. In order to meet this challenge, this paper proposes an improved method Mutual Information Domain-Adversial Neural Networks (DANN) based on domain adaptation, which improves the domain discriminator to better extract domain invariant features. In addition, to avoid the loss of target information, the mutual information among target features, source features, and original target data is calculated to fix the features on the target site during the migration process. Different from the traditional methods, we only use 40% of the data sets for training, and the rest are used for prediction, so we can complete the prediction of more scenes. Experimental results show that this method can accurately and stably predict SOH.
In this paper, a fractional order sliding mode guidance law to stationary targets is proposed which can also achieve impact angle constraint. A new independent variable is introduced. The sliding surface is well analyzed by both Lyapunov stability and analytical expressions. The fractional order term and initializing term affect the variations of the sliding surface, flight path angle and trajectory. Regulations are summarized according to the guidance expectations. By setting parameters properly, the fractional order guidance law has further hit downrange compared with the integer one. The extension of hit downrange can be applied in practical engineering. The proposed guidance law shows good effectiveness and robustness.
This paper presents an improved approach called error dynamics based dual heuristic dynamic programming (ED-DHP) for nonlinear flight attitude control. The augmented system state containing tracking error and reference trajectory are defined for the DHP to solve the tracking problem, without requiring the complete knowledge of the vehicle dynamics and the reference trajectory dynamics. A recursive least square method is used to identify not only the input dynamics, but also the total uncertainty of the system based on the error dynamics in real time. The simulations of two different air vehicles demonstrate the good performance of ED-DHP for the online self-learning flight attitude control.
To accurately predict the nitrogen oxides (NOx) emissions concentration during the municipal solid waste incineration process, a prediction method based on the Bagging ensemble and stochastic configuration network (SCN) is proposed, and the relevant variables of the municipal solid waste incineration process are used to predict NOx concentration in this paper. Firstly, the Bootstrap sampling method is used to generate several different training subsets, and multiple SCN base models are trained under different subsets. The average value of the output results of every base model is taken as the final output. Finally, the actual historical data of a solid waste treatment plant in Beijing are used to verify the model and compared with the single SCN, random vector function link network (RVFLN), Bagging-RVFLN. The experimental results show that the proposed method has high accuracy and can accurately predict the NOx emissions concentration in MSW incineration process.
This paper investigates the appointed-time prescribed performance attitude tracking control problem of a rigid spacecraft using the adaptive critic control. Firstly, the error transformation technique is utilized to convert the system with the prescribed performance constraint into an equivalent unconstraint optimal control problem. Then in the framework of adaptive critics, the critic-only architecture is built to solve the Hamilton-Jacobi-Bellman equations associated with these solutions. A rigorous mathematical stability proof is given. Finally, numerical simulations are presented to demonstrate the effectiveness and robustness of the proposed controller.
In this paper, the fault-tolerant control problem is investigated for AUV with thruster faults and thruster saturation. A non-singular terminal sliding mode control method is developed based on a third-order fast finite time extended state observer (FFTESO). Firstly, the third-order FFTESO is introduced to estimate the generalized uncertainty, which includes current disturbance, dynamic model uncertainty, and thruster fault and saturation. Then, a non-singular terminal sliding mode-based fault tolerant control is presented by combining the estimation provided by the observer. And the Nussbaum-type function is applied to deal with the unknown gain caused by the thruster saturation. It is proved theoretically that the stability of the observer and the designed fault-tolerant controller can ensure the boundness of tracking errors. Finally, the effectiveness of the proposed control scheme is evaluated by simulation and comparison experiments on ODIN-AUV.
The distributed flow shop group scheduling problem (DFGSP) has a wide practical application, especially in the cellular manufacturing field. Therefore, for the above complexity of the problem, this article proposed a mathematical model of DFGSP with the makespan criterion and a modified iterated greedy algorithm (MIG), including global search and local enhancement strategies. We have done numerous experiments on 810 instances to verify the effectiveness of the proposed MIG.
An adaptive cubature Kalman filter is proposed in this paper to estimate the state of charge (SOC) of a lithium-ion battery. Firstly, a second-order RC equivalent circuit model is constructed to describe the dynamic characteristics of a lithium-ion battery. Secondly, we design a linear Kalman filter and a cubature Kalman filter to achieve the adaptive estimation of SOC, model parameters and the coefficients in the measurement equation. Thirdly, the noise covariance matrices are adaptively adjusted in order to further improve the estimation accuracy of the proposed algorithm. Finally, the estimation accuracy and adaptability of the algorithm proposed in this paper are verified by different experiments.
This paper addresses a class of optimization problems with time-varying cost functions by proposing a fully distributed prescribed-time algorithm. The algorithm decomposes the overall optimization problem into three successive subproblems, which are solved sequentially. During the three stages of the algorithm, the estimation of the total cost function's average gradient information, consensus among the states, and tracking of the optimal state trajectories are achieved in turn. Given the segmentation strategy's demand for rapid convergence, the algorithm ensures convergence within a prescribed time. Using the Lyapunov method, it is shown that all three subproblems can be solved within any user-prescribed time, independent of the system's initial states or topology. To further exploit the independence of prescribed-time convergence from system states, the algorithm eliminates the reliance on system topology information in parameter settings by introducing adaptive parameters in place of traditional fixed ones, thus enabling fully distributed control. Finally, numerical simulations and an UAV target tracking experiment are conducted to validate the effectiveness and practicality of the proposed algorithm.
Today’s industrial production models are becoming increasingly complex. Jobs of the same type are usually divided into groups for machining. Factories are often designed in a distributed structure, i.e., they are built in serval different locations. The distributed flowshop group scheduling problem with blocking constraints (DBFGSP) has not been well studied. We constructed a mixed-integer linear programming model, and an iterated greedy algorithm based on the accelerated calculation (IGACA) to reduce the time complexity. We use 810 instances to compare our proposed algorithm with four other excellent algorithms used to solve related problems. The experimental result shows that our proposed algorithm is advantageous in solving the DBFGSP.
A path planning model based on the A * algorithm and improved ant colony algorithm is proposed to solve the path optimization problem of handling robot sorting multiple goods at one time. Firstly, the plan of the warehouse and the location of the goods are converted into black and white grid maps by grid method, and then the A * algorithm is used to convert the Manhattan distance between all nodes into an undirected graph. Secondly, the pheromone updating principle of the traditional ant colony algorithm is optimized to artificially release additional pheromones on the optimal path to enhance the positive feedback effect, thus speeding up its convergence speed and obtaining the optimal solution. Finally, the performance of the algorithm is tested by simulation. The simulation results show that compared with the traditional ant colony algorithm, the improved ant colony algorithm has better convergence speed and higher efficiency under the condition of a large number of nodes.
Aiming at the problems of high coupling, parameter uncertainty and external environment interference in practical applications of six degrees of freedom manipulator, a control strategy based on improved active disturbance rejection control was proposed. Firstly, in order to improve the response speed of the system, the sliding mode control tracking differentiator (SMC-TD) is designed based on the synovial theory.In order to reduce the system jitter, the new sliding mode function is used to improve the sliding mode control law, which reduces the system jitter and improves the system response speed. Secondly, the han function is constructed to realize the design of continuous smooth extended state observer, and the improved fractional extended state observer (FO-ESO) is designed in combination with fractional calculus, which enhances the anti-interference ability of the system. Finally, combined with the traditional nonlinear error feedback control law (NLSEF), an improved active disturbance rejection controller was built, and different control strategies were used to compare the six degrees of freedom manipulator in the same environment. The experimental results show that the controller has good control effect and anti-disturbance robustness, which proves the effectiveness of the proposed method.
Sentiment Analysis is an important task in NLP. In recent years, more and more attention has been paid to Multimodal Sentiment Analysis. For Multimodal Aspect-Based Sentiment Analysis (MABSA), most of the existing methods take the original information of each modality as input and extract each modality’s features and the alignment between them. Based on the existing methods, additional information, including aspect terms and image caption, is extracted. The original and additional information is used as a compound feature so that the model can have a better understanding of each modality and the relationship between them. Experiments show that the method achieves better results on MABSA and its two sub-tasks: Multimodal Aspect-Sentiment Classification (MASC) and Multimodal Aspect-Terms Extraction (MATE).
