自动化学报最新文献

The linguistic dynamic systems (LDSs) based on type-1 fuzzy sets can provide a powerful tool for modeling, analysis, evaluation and control of complex systems. However, as pointed out in earlier studies, it is much more reasonable to take type-2 fuzzy sets to model the existing uncertainties of linguistic words. In this paper, the LDS based on type-2 fuzzy sets is studied, and its reasoning process is realized through the perceptual reasoning method. The properties of the perceptual reasoning method based LDS (PR-LDS) are explored. These properties demonstrated that the output of PR-LDS is intuitive and the computation complexity can be reduced when the consequent type-2 fuzzy numbers in the rule base satisfy some conditions. Further, a data driven method for the design of the PR-LDS is provided. At last, the effectiveness and rationality of the proposed data-driven method are verified by an example.

This paper proposes a sensor fault diagnosis method for a class of discrete-time linear time-varying (LTV) systems. In this paper, the considered system is firstly formulated as a descriptor system representation by considering the sensor faults as auxiliary state variables. Based on the descriptor system model, a fault estimation filter which can simultaneously estimate the state and the sensor fault magnitudes is designed via a minimum-variance principle. Then, a fault diagnosis scheme is presented by using a bank of the proposed fault estimation filters. The novelty of this paper lies in developing a sensor fault diagnosis method for discrete LTV systems without any assumption on the dynamic of fault. Another advantage of the proposed method is its ability to detect, isolate and estimate sensor faults in the presence of process noise and measurement noise. Simulation results are given to illustrate the effectiveness of the proposed method.

The box constraints in image restoration have been arousing great attention, since the pixels of a digital image can attain only a finite number of values in a given dynamic range. This paper studies the box-constrained total-variation (TV) image restoration problem with automatic regularization parameter estimation. By adopting the variable splitting technique and introducing some auxiliary variables, the box-constrained TV minimization problem is decomposed into a sequence of subproblems which are easier to solve. Then the alternating direction method (ADM) is adopted to solve the related subproblems. By means of Morozov's discrepancy principle, the regularization parameter can be updated adaptively in a closed form in each iteration. Image restoration experiments indicate that with our strategies, more accurate solutions are achieved, especially for image with high percentage of pixel values lying on the boundary of the given dynamic range.

A graph-based multiple view line matching method is proposed based on results of multiple view stereo (MVS) algorithms. With the 3D points and their visibility information provided by MVS, point-line correspondences are firstly established through 3D-to-2D re-projection. Each image line detected in different views is described using a 3D point set as well as a unit vector representing its coarse 3D direction. From such a description, pairwise similarity and consistency are evaluated. Then, a graph is constructed to contain all image lines as nodes. To get a unified node distance measure, a spectral graph analysis method is employed. Finally, a modified DBSCAN algorithm is introduced to obtain reliable line matches from the graph. Experiments show that our method is more robust and exhibits better accuracy than the existing methods.

It is promising to improve web image search results through exploiting the results' visual contents for learning a binary classifier which is used to refine the results' relevance degrees to the given query. This paper proposes an algorithm framework as a solution to this problem and investigates the key issue of training data selection under the framework. The training data selection process is divided into two stages: initial selection for triggering the classifier learning and dynamic selection in the iterations of classifier learning. We investigate two main ways of initial training data selection, including clustering based and ranking based, and compare automatic training data selection schemes with manual manner. Furthermore, support vector machines and the max-min pseudo-probability (MMP) based Bayesian classifier are employed to support image classification, respectively. By varying these factors in the framework, we implement eight algorithms and tested them on keyword based image search results from Google search engine. The experimental results confirm that how to select the training data from noisy search results is really a key issue in the problem considered in this paper and show that the proposed algorithm is effective to improve Google search results, especially at top ranks, thus is helpful to reduce the user labor in finding the desired images by browsing the ranking in depth. Even so, it is still worth meditative to make automatic training data selection scheme better towards perfect human annotation.

A novel approach based on a type of simplified motion planning (SMP) is presented in this paper to generate online trajectory for manipulator systems with multiple degrees of freedom (DOFs). The key issue is to find minimum-acceleration trajectory planning (MATP) to optimize the arm motion to reduce disturbance. Moreover, necessary and sufficient conditions for solution's existence subject to all the kinematic constraints of joint position, velocity, acceleration and jerk are devised. Besides, this new method can be activated online from the arbitrary initial state to the arbitrary target state so that it enables the robot to change the original path at any time. Finally, the approach is applied to a real humanoid robot arm with seven DOFs to show its efficiency.

This paper is concerned with statistical χ² testing based fault detection (FD) for a class of linear discrete time-varying (LDTV) stochastic systems with delayed state. Different from the traditional residual based FD, we propose to construct the evaluation function by directly using measurement observations. Then an equivalent solution can be given in terms of Riccati recursion by utilizing projection and innovation analysis technique. Moreover, the fault free case evaluation function is with central χ² distribution and the heavy computational burden is reduced. Furthermore, strategies of χ² statistic testing on evaluation function are also discussed. Finally, a numerical example is given to illustrate the proposed method.

This paper presents a simple and explicit method for G²-constrained merging of a pair of Bézier curves by minimizing the l₂ distance defined in terms of control points. After expressing the l₂ distance as a quadratic function of two parameters, the optimally merged curve can be explicitly obtained, which is achieved by control point optimization such that the l₂ distance is minimized. The existence of the unique solution is shown by proving that the l₂ distance is convex. The proposed method is explicit and efficient since it is non-iterative and expressed by known control points. Numerical examples demonstrate the effectiveness of the new method.

The convergence analysis of the standard particle swarm optimization (PSO) has shown that the changing of random functions, personal best and group best has the potential to improve the performance of the PSO. In this paper, a novel strategy with variable random functions and polynomial mutation is introduced into the PSO, which is called particle swarm optimization algorithm with variable random functions and mutation (PSO-RM). Random functions are adjusted with the density of the population so as to manipulate the weight of cognition part and social part. Mutation is executed on both personal best particle and group best particle to explore new areas. Experiment results have demonstrated the effectiveness of the strategy.