This paper aims to increase the Unmanned Aerial Vehicle's (UAV) capacity for target tracking. First, a control model based on fuzzy logic is created, which modifies the UAV's flight attitude in response to the...
详细信息
This paper aims to increase the Unmanned Aerial Vehicle's (UAV) capacity for target tracking. First, a control model based on fuzzy logic is created, which modifies the UAV's flight attitude in response to the target's motion status and changes in the surrounding environment. Then, an edge computing-based target tracking framework is created. By deploying edge devices around the UAV, the calculation of target recognition and position prediction is transferred from the central processing unit to the edge nodes. Finally, the latest Vision Transformer model is adopted for target recognition, the image is divided into uniform blocks, and then the attention mechanism is used to capture the relationship between different blocks to realize real-time image analysis. To anticipate the position, the particle filter algorithm is used with historical data and sensor inputs to produce a high-precision estimate of the target position. The experimental results in different scenes show that the average target capture time of the algorithm based on fuzzy logic control is shortened by 20% compared with the traditional proportional-integral-derivative (PID) method, from 5.2 s of the traditional PID to 4.2 s. The average tracking error is reduced by 15%, from 0.8 m of traditional PID to 0.68 m. Meanwhile, in the case of environmental change and target motion change, this algorithm shows better robustness, and the fluctuation range of tracking error is only half of that of traditional PID. This shows that the fuzzy logic control theory is successfully applied to the UAV target tracking field, which proves the effectiveness of this method in improving the target tracking performance.
In order to solve the fusion estimation problem of multi-sensor with unknown cross-covariance,an improved suboptimal fusion algorithm weighted by matrices is proposed for nonlinear ***,for significance of linear minim...
详细信息
ISBN:
(数字)9789887581581
ISBN:
(纸本)9798350366907
In order to solve the fusion estimation problem of multi-sensor with unknown cross-covariance,an improved suboptimal fusion algorithm weighted by matrices is proposed for nonlinear ***,for significance of linear minimum variance,the simplest constraints based on fusion weighted by matrices are derived by Shure complement *** constraints can ensure the positive definiteness of the fusion estimate error covariance,and the consistency of the proposed suboptimal fusion ***,a suboptimal fusion estimation weighted by matrices is proposed based on linear matrix inequality(LMI).Considering the time-consuming problem in the optimization process of LMI algorithm and the complexity of the nonlinear system,the optimal value is obtained by the nonlinear auto-regressive neural network with exogenous input(NARX).Finally,a nonlinear suboptimal fusion algorithm weighted by matrices based on LMI and NARX is proposed in combination with the particle filter algorithm(PF).
Micro milling aims to manufacture miniature structures with high quality and complex features, and the stochastic time-varying tool wear is a crucial factor which has great influence on machining quality and efficienc...
详细信息
Micro milling aims to manufacture miniature structures with high quality and complex features, and the stochastic time-varying tool wear is a crucial factor which has great influence on machining quality and efficiency of micro milling process. To improve the precision of machining and sustainability of micro cutting tools, the in-process tool wear conditions should be identified and updated ahead of time. In this work, an improved integrated estimation method is proposed based on the long short-term memory (LSTM) network and particlefilter (PF) algorithm to predict the stochastic tool wear values. The integrated PF-LSTM identification methodology is developed to predict the in-process stochastic tool wear progression on the basis of the historical measurement data. With the estimation of in-process stochastic tool wear, the cutting force model is modified, in which the influence of tool run-out and the trochoidal trajectory of cutting edge are also considered. The proposed integrated estimation method of in-process stochastic tool wear and the modified cutting force model were validated by the micro milling experiments with workpiece material Al6061. It can be seen from the comparison results that the availability and sustainability of micro cutting tool have been improved, and the prediction accuracy also could be increased by 3.4% compared with that without considering the influence of tool wear.
This study incorporates the Markov switching model with return jumps to depict the behavior of stock returns. Based on the daily Standard & Poor's 500 index (hereafter SPX) and the daily closing price of the c...
详细信息
This study incorporates the Markov switching model with return jumps to depict the behavior of stock returns. Based on the daily Standard & Poor's 500 index (hereafter SPX) and the daily closing price of the call option, we use the particlefiltering algorithm to fit the parameter of the model. The joint log-likelihood evaluates the model performance: the weighted average log-likelihood with the rate of return of the SPX and the relative implied volatility root-mean-squared error for the SPX call options. The empirical results identify that the pricing model with jump risks improves the pricing performance to the median-term call options. According to the sensitivity analysis, option prices increase with the probability of remaining in the recession state but decrease with the probability of remaining in the expansion state. Moreover, the call option prices are positively associated with the volatility in each market state and the factors of jump risk.
With lithium-ion batteries are more and more widely used in transportation, the estimation of battery state-of-health（SOH） is of great significance in the safe and reliable operation of battery management system and...
详细信息
With lithium-ion batteries are more and more widely used in transportation, the estimation of battery state-of-health（SOH） is of great significance in the safe and reliable operation of battery management system and the reduction of maintenance cost. Based on the analysis of the traditional particle filter algorithm, the genetic factor of genetic algorithm（GA）is introduced into the particlefilter and improved by adaptive mutation. In order to predict the SOH of lithium-ion battery, the health index（HI） is extracted from the measurable parameters of lithium-ion battery. The mapping model between HI index and SOH is established and applied to the observation of state space model. In this paper, a battery SOH estimation method based on improved particle filter algorithm is proposed. The experimental results show that the proposed method is superior to the traditional particlefilter（PF） algorithm and has good accuracy in estimating the degradation process of lithium-ion batteries.
With the rapid development of e-commerce, the security issues of collaborative filtering recommender systems have been widely investigated. Malicious users can benefit from injecting a great quantities of fake profile...
详细信息
With the rapid development of e-commerce, the security issues of collaborative filtering recommender systems have been widely investigated. Malicious users can benefit from injecting a great quantities of fake profiles into recommender systems to manipulate recommendation results. As one of the most important attack methods in recommender systems, the shilling attack has been paid considerable attention, especially to its model and the way to detect it. Among them, the loose version of Group Shilling Attack Generation algorithm (GSAGenl) has outstanding performance. It can be immune to some PCC (Pearson Correlation Coefficient)-based detectors due to the nature of anti-Pearson correlation. In order to overcome the vulnerabilities caused by GSAGenl, a gravitation-based detection model (GBDM) is presented, integrated with a sophisticated gravitational detector and a decider. And meanwhile two new basic attributes and a particle filter algorithm are used for tracking prediction. And then, whether an attack occurs can be judged according to the law of universal gravitation in decision-making. The detection performances of GBDM, HHT-SVM, UnRAP, AP-UnRAP Semi-SAD,SVM-TIA and PCA-P are compared and evaluated. And simulation results show the effectiveness and availability of GBDM.
Multipath interference is the main error source for high-precision positioning applications, and multipath suppression and mitigation is particularly important. particlefilters are widely used to solve non-linear fil...
详细信息
Multipath interference is the main error source for high-precision positioning applications, and multipath suppression and mitigation is particularly important. particlefilters are widely used to solve non-linear filtering problems without limitation of Gaussian distribution, and Global navigation satellite system (GNSS) multipath estimation and mitigation based on particlefilter are proposed in this study. This approach has four innovations: Firstly, the Kalman-based multipath signal model is improved to obtain a particlefilter multipath signal model, and I and Q two channel signals in the algorithm model solve the coherence correlation phase estimation problem. Secondly, the particle filter algorithm constructs the correlation function using the current particle amplitude and delay. Thirdly, taking full advantage of every particle information, the weighted average is used to calculate state quantity of multipath. Fourthly, three resampling algorithms, including simple random, pseudo-parallel genetic algorithm and niche genetic algorithm, are taken to resample. particle filter algorithm for multipath estimation and mitigation based on simulated data and actual navigation satellite signal data is verified. The simulation results show the proposed algorithm has high accuracy in multipath estimation than MEDLL and TK-MEDLL, especially in multichannel multipath estimation. The actual experimental results show particle filter algorithm is effective for improving the positioning accuracy in complex environments.
The particle degradation problem of particlefilter (PF) algorithm caused by reduction of particle weights significantly influences the positioning accuracy of target nodes in wireless sensor networks. This study pres...
详细信息
The particle degradation problem of particlefilter (PF) algorithm caused by reduction of particle weights significantly influences the positioning accuracy of target nodes in wireless sensor networks. This study presents a predictor to obtain the particle swarm of high quality by calculating non-linear variations of ranging between particles and flags and modifying the reference distribution function. To this end, probability variations of distances between particles and star flags are calculated and the maximum inclusive distance using the maximum probability of high-quality particle swarm is obtained. The quality of particles is valued by the Euclidean distance between the predicted and real observations, and hereafter particles of high quality are contained in spherical coordinate system using the distance as diameter. The simulation results show that the proposed algorithm is robust and the computational complexity is low. The method can effectively improve the positioning accuracy and reduce the positioning error of target nodes.
The absolute positioning accuracy of the industrial robot is one of its important performance indexes, which is impacted by the key factor of robotic kinematic parameters. Therefore, based on the MDH model a calibrati...
详细信息
ISBN:
(纸本)9781728158556
The absolute positioning accuracy of the industrial robot is one of its important performance indexes, which is impacted by the key factor of robotic kinematic parameters. Therefore, based on the MDH model a calibration method of robot kinematic parameters, which combines the Levenberg-Marquardt algorithm with the particle filter algorithm is proposed. Firstly, the MDH model of an industrial robot is established, and the parameters in the tool coordinate transformation are also regarded as the parameters to be calibrated. Then the end error model is established. Secondly, the initial optimization is carried out using the Levenberg-Marquardt (LM) algorithm. Meanwhile, the particle filter algorithm is used to further optimize the parameters considering the measurement error. Finally, compared with other methods, such as spatial circle fitting method, least square method and extended Kalman filteralgorithm. The results show that the kinematic parameters of the robot are accurately calibrated and the absolute positioning accuracy of the industrial robot is significantly improved by this method. Compared with other methods, the parameters calibrated by this method have stronger generalization ability.
The article discusses the suitability of the multiwall radio wave propagation model for RSSI reference data preparation for fingerprinting-based indoor positioning applications. Localization system employs Bluetooth L...
详细信息
暂无评论