This paper focuses on an accelerating method for partitioning the loops in the structure of the adaptive dynamic programming(ADP). ADP contains critic-actor structure which involves the iterations of the value funct...
详细信息
This paper focuses on an accelerating method for partitioning the loops in the structure of the adaptive dynamic programming(ADP). ADP contains critic-actor structure which involves the iterations of the value function. When the system needs to be stable, the value function generally needs to iterate thousands of times, the high computation burden which hinders the iterations will be generated. In order to reduce the computation burden, we introduce a hyperparallelepiped based loop partitioning(H-LP) method which splits the iterations of the value function and reduces the communication traffic calculated by the data footprint. The experiment results show that the computation performance will be enhanced when the H-LP method is introduced. The proposed method has an important practical significance.
This paper presents a position control strategy based on the iterative method for a planar *** control objective of the system is to move the end-point from any initial equilibrium point to a target equilibrium *** pr...
详细信息
This paper presents a position control strategy based on the iterative method for a planar *** control objective of the system is to move the end-point from any initial equilibrium point to a target equilibrium *** presented method is based on the iterative steering,where a converging control law is applied *** order to compute such a control law,the dynamic equations of the system are transformed via partial feedback linearization and nilpotent ***,the simulation results demonstrate that the position control objective is realized by using this control strategy.
An important feature of the deep learning algorithm is that the hidden layer of the neural network is more dependent on more computing resources and a larger amount of data. In this paper, we use Triplet GAN method to...
详细信息
An important feature of the deep learning algorithm is that the hidden layer of the neural network is more dependent on more computing resources and a larger amount of data. In this paper, we use Triplet GAN method to identify human body images with tattoos using fewer data sample labels. The model achieved better results(0.9462) than the only Triplet(0.8352) and only GAN(0.9178) in the MNIST dataset, it also achieved higher recognition accuracy(77-82%) under the premise of saving the amount of data, compared with the traditional machine learning algorithm(54-84%). We provide a new idea for tattoo image recognition applications in embedded computing units.
In the process of image acquisition and transmission, the image always generates noise due to internal and external interference. Noise reduces the quality of the image, and makes it difficult for subsequent image pro...
详细信息
In the process of image acquisition and transmission, the image always generates noise due to internal and external interference. Noise reduces the quality of the image, and makes it difficult for subsequent image processing. Therefore, image denoising is very important in image processing. Wavelet denoising can effectively filter out noise and retain high-frequency information of the image, this method has the characteristics of fast operation speed and has become an important branch of image denoising. Threshold functions commonly used in wavelet threshold denoising include hard threshold function and soft threshold function. The hard threshold function is not continuous as a whole. Although the soft threshold function has good continuity, there is always a constant deviation between the processed coefficient and the original coefficient when the wavelet coefficient is large. In response to these deficiencies, this paper establishes a new improved threshold function based on traditional soft and hard threshold functions. By processing the thresholds of wavelet coefficients, a reasonable balance between smoothing and edge oscillations can be achieved after image denoising. The improved threshold function not only overcomes the shortcomings of the soft and hard threshold functions, but also provides more flexibility in the processing of image *** MATLAB simulation, the denoising effects of the soft, hard threshold functions and the threshold function constructed in this paper are compared in terms of signal-to-noise ratio(SNR) and root mean square error(MSE). The MATLAB simulation results show that compared with the traditional threshold function, the improved threshold function has a higher signal-to-noise ratio(SNR = 26.27709) and a smaller mean square error(MSE = 153.4579), and it has a good noise reduction effect.
The human-computer interaction technology which based on the gaze tracking system is convenient and fast. It can achieve the purpose of sight and computer interaction. Based on the relatively static head tracking syst...
详细信息
The human-computer interaction technology which based on the gaze tracking system is convenient and fast. It can achieve the purpose of sight and computer interaction. Based on the relatively static head tracking system, we make improvements for OTSU algorithm to separate the binarized pupil image and the background image completely. I wrote a function to delete a small area that gets a complete and clear binarized pupil image for the existence of small area noise clump. Finally, using the contour extraction method and the three-point circle method to complete the task of Pupil location.
Maximum power point tracking controller is essential to obtain the maximum power from a solar array in the photovoltaic systems as the PV power module varies with the temperature and solar irradiation. In the DC/DC ci...
详细信息
Maximum power point tracking controller is essential to obtain the maximum power from a solar array in the photovoltaic systems as the PV power module varies with the temperature and solar irradiation. In the DC/DC circuit, the maximum power point tracking algorithm based on parabolic approximation method is used. On the basis of analyzing the principle of various tracking methods, the key technology of parabola approximation can be found to find the exact maximum power point.
A position control strategy by employing the model reduction and the cascade transformation is proposed for a planar four-link AAPA(Active-Active-Passive-Active) underactuated manipulator in this ***,a dynamics model ...
详细信息
A position control strategy by employing the model reduction and the cascade transformation is proposed for a planar four-link AAPA(Active-Active-Passive-Active) underactuated manipulator in this ***,a dynamics model of the system is ***,three controllers are designed based on the Lyapunov function to control the fourth active link from any initial angle to zero while the angles of the first and second active link remain their initial values,which makes the system is reduced to a planar virtual three-link AAP(Active-Active-Passive) underactuated ***,we obtain the new inputs of the cascade system of the planar virtual three-link AAP underactuated ***,we can get the controllers of the active links to realize the system position control *** results demonstrate the validity of the proposed control strategy.
To ensure the efficiency of people’s work while working long hours,an initiative service method with regard to degree of sleepiness for drinking service robot is *** can recognize the degree of sleepiness and provide...
详细信息
To ensure the efficiency of people’s work while working long hours,an initiative service method with regard to degree of sleepiness for drinking service robot is *** can recognize the degree of sleepiness and provide initiative service to increase the efficiency of people’s *** of sleepiness is defined as three levels by observer rating,which is supposed to be associated with human actions,environmental factors,and individual *** addition,a model of degree of sleepiness based on Finite State Machines(FSM) is *** relationship between human demands and degree of sleepiness is established through demand analysis model by which the initiative service with regard to degree of sleepiness for drinking service robot is *** on real initiative drinking service are performed in a multimodal emotional communication based on human-robot interaction(MEC-HRI) system,from which the experimental results show that the time of initiative service is 8 minutes earlier than that without initiative *** prospect,the proposal could be applied to other kinds of services such as health care,safety driving assistance,and other safety work assistance.
As slide steering technology has lower maintenance costs, it is widely used in geological drilling industry. In order to adjust the hole trajectory, this technology changes the drilling direction by controlling tool f...
详细信息
As slide steering technology has lower maintenance costs, it is widely used in geological drilling industry. In order to adjust the hole trajectory, this technology changes the drilling direction by controlling tool face angle of downhole power drill tool. However, due to the existence of the untwist angle, it is difficult to precisely control the angle, which will directly affect the quality of hole trajectory. So untwist angle prediction is the prerequisite of hole trajectory control. This paper introduces a common method for calculating untwist angle for generating the training set. And then factors that influence untwist angle will be analyzed. Meanwhile, based on the analysis and calculation results, support vector regression is introduced in the prediction algorithm to provide a new way for untwist angle prediction.
A demand analysis method based on TAKAGI-SUGENO(T-S) fuzzy model for drinking service is proposed to provide corresponding services according to users’ emotions and intentions in human-robot interaction,in which T-...
详细信息
A demand analysis method based on TAKAGI-SUGENO(T-S) fuzzy model for drinking service is proposed to provide corresponding services according to users’ emotions and intentions in human-robot interaction,in which T-S fuzzy model is used to establish the relationship among human intention and human ***,the transformation of input and output is ***,fuzzy rules are formulated,and then fuzzy inference is applied to get user’s demand corresponding to emotion and *** proposal considers peoples fuzziness in inferring humans intention,which could help the robots to provide satisfied drinking service to *** validate the proposal,drinking service experiments are performed in a laboratory scenario using a humans-robots interaction system,from which the experimental results demonstrate the feasibility of the proposal.
暂无评论