A new method for the direct adaptive regulation of unknown nonlinear dynamical systems is proposed in this paper, paying special attention to the analysis of the model order problem. The method uses a neuro-fuzzy (NF) modeling of the unknown system, which combines fuzzy systems (FSs) with high order neural networks (HONNs). We propose the approximation of the unknown system by a special form of an NF-dynamical system (NFDS), which, however, may assume a smaller number of states than the original unknown model. The omission of states, referred to as a model order problem, is modeled by introducing a disturbance term in the approximating equations. The development is combined with a sensitivity analysis of the closed loop and provides a comprehensive and rigorous analysis of the stability properties. An adaptive modification method, termed ‘parameter hopping’, is incorporated into the weight estimation algorithm so that the existence and boundedness of the control signal are always assured. The applicability and potency of the method are tested by simulations on well known benchmarks such as ‘DC motor’ and ‘Lorenz system’, where it is shown that it performs quite well under a reduced model order assumption. Moreover, the proposed NF approach is shown to outperform simple recurrent high order neural networks (RHONNs).

Interest in inverse reinforcement learning (IRL) has recently increased, that is, interest in the problem of recovering the reward function underlying a Markov decision process (MDP) given the dynamics of the system and the behavior of an expert. This paper deals with an incremental approach to online IRL. First, the convergence property of the incremental method for the IRL problem was investigated, and the bounds of both the mistake number during the learning process and regret were provided by using a detailed proof. Then an online algorithm based on incremental error correcting was derived to deal with the IRL problem. The key idea is to add an increment to the current reward estimate each time an action mismatch occurs. This leads to an estimate that approaches a target optimal value. The proposed method was tested in a driving simulation experiment and found to be able to efficiently recover an adequate reward function.

This paper deals with Wiener model based predictive control of a pH neutralization process. The dynamic linear block of the Wiener model is parameterized using Laguerre filters while the nonlinear block is constructed using least squares support vector machines (LSSVM). Input-output data from the first principle model of the pH neutralization process are used for the Wiener model identification. Simulation results show that the proposed Wiener model has higher prediction accuracy than Laguerre-support vector regression (SVR) Wiener models, Laguerre-polynomial Wiener models, and linear Laguerre models. The identified Wiener model is used here for nonlinear model predictive control (NMPC) of the pH neutralization process. The set-point tracking performance of the proposed NMPC is compared with those of the Laguerre-SVR Wiener model based NMPC, Laguerre-polynomial Wiener model based NMPC, and linear model predictive control (LMPC). Validation results show that the proposed NMPC outperforms the other three controllers.

Traditional cartoons have been widely used in entertainment, education, and advertisement. Thus, a large amount of cartoon data is available. In this paper, we propose a new technique for capturing the motion of a character in an existing cartoon sequence. This technique tracks the contours of the cartoon character in the sequence, and key frames are used to guide the tracking. We model contour tracking as a space-time optimization problem in which an energy function including both temporal and spatial constraints is defined. First, the user labels the contours of the character on the key frames. Then, the contours on the intermediate frames are tracked by minimizing the energy function. The user may need to interactively adjust the tracking result and restart the optimization process to refine the result. Finally, an edge snapping algorithm is applied to make the tracking result more precise. Experiments show that our technique works effectively.

To provide pest technicians with a convenient way to recognize insects, a novel method is proposed to classify insect images by integrated region matching (IRM) and dual tree complex wavelet transform (DTCWT). The wing image of the lepidopteran insect is preprocessed to obtain the region of interest (ROI) whose position is then calibrated. The ROI is first segmented with the k-means algorithm into regions according to the color features, properties of all the segmented regions being used as a coarse level feature. The color image is then converted to a grayscale image, where DTCWT features are extracted as a fine level feature. The IRM scheme is undertaken to find K nearest neighbors (KNNs), out of which the nearest neighbor is searched by computing the Canberra distance of DTCWT features. The method was tested with a database including 100 lepidopteran insect species from 18 families and the recognition accuracy was 84.47%. For the forewing subset, a recognition accuracy of 92.38% was achieved. The results showed that the proposed method can effectively solve the problem of automatic species identification of lepidopteran specimens.

A new separation algorithm based on contour segments and ellipse fitting is proposed to separate the ellipse-like touching grain kernels in digital images. The image is filtered and converted into a binary image first. Then the contour of touching grain kernels is extracted and divided into contour segments (CS) with the concave points on it. The next step is to merge the contour segments, which is the main contribution of this work. The distance measurement (DM) and deviation error measurement (DEM) are proposed to test whether the contour segments pertain to the same kernel or not. If they pass the measurement and judgment, they are merged as a new segment. Finally with these newly merged contour segments, the ellipses are fitted as the representative ellipses for touching kernels. To verify the proposed algorithm, six different kinds of Korean grains were tested. Experimental results showed that the proposed method is efficient and accurate for the separation of the touching grain kernels.

In multi-criterion decision making applications, a skyline query narrows the search range, as it returns only the points that are not dominated by others. Unfortunately, in high-dimensional/large-cardinal datasets there exist too many skyline points to offer interesting insights. In this paper, we propose a novel structure, called the dominance tree (Do-Tree), to effectively index and retrieve the skyline. Do-Tree is a straightforward and flexible tree structure, in which skyline points are resident on leaf nodes, while the internal nodes contain the entries that dominate their children. As Do-Tree is built on a dominance relationship, it is suitable for the retrieval of specified skyline via dominance-based predicates customized by users. We discuss the topology of Do-Tree and propose the construction methods. We also present the scan scheme of Do-Tree and some useful queries based on it. Extensive experiments confirm that Do-Tree is an efficient and scalable index structure for the skyline.

We present a novel pipelined fast Fourier transform (FFT) architecture which is capable of producing the output sequence in normal order. A single-path delay commutator processing element (SDC PE) has been proposed for the first time. It saves a complex adder compared with the typical radix-2 butterfly unit. The new pipelined architecture can be built using the proposed processing element. The proposed architecture can lead to 100% hardware utilization and 50% reduction in the overall number of adders required in the conventional pipelined FFT designs. In order to produce the output sequence in normal order, we also present a bit reverser, which can achieve a 50% reduction in memory usage.