Search

In order to overcome the influence of internal and external disturbances caused by rotor tilt motion and gust disturbance on the full flight mode control of a tilt-rotor unmanned aerial vehicle (UAV), a design method using fuzzy backstepping control based on an extended-state observer (FBS-ESO) is proposed. In this paper, fuzzy control is used to tune the parameters of the backstepping control law online, and the extended-state observer estimates the total disturbance of the controlled system to improve the controller’s robustness and anti-disturbance capability. This paper designs the attitude control system of a tilt-rotor UAV based on an FBS-ESO controller. The control performance of the FBS-ESO controller is tested in a hardware-in-loop simulation of the attitude control system. The simulation results show that changing the rotor tilt angle will destroy the stability of the traditional backstepping controller and active disturbance rejection controller (ADRC). In contrast, the FBS-ESO controller maintains good control performance. In addition, the performance of the FBS-ESO controller is not be significantly affected by adding external gust disturbance or changing the UAV parameters in the simulation. These disturbances significantly impact the traditional backstepping controller and ADRC. Therefore, the FBS-ESO controller has better anti-disturbance capabilities and robustness, as well as higher attitude control accuracy.

In this paper, we will investigate the contribution of color names for the task of salient object detection. An input image is first converted to color name space, which is consisted of 11 probabilistic channels. By exploiting a surroundedness cue, we obtain a saliency map through a linear combination of a set of sequential attention maps. To overcome the limitation of only using the surroundedness cue, two global cues with respect to color names are invoked to guide the computation of a weighted saliency map. Finally, we integrate the above two saliency maps into a unified framework to generate the final result. In addition, an improved post-processing procedure is introduced to effectively suppress image backgrounds while uniformly highlight salient objects. Experimental results show that the proposed model produces more accurate saliency maps and performs well against twenty-one saliency models in terms of three evaluation metrics on three public data sets., http://www.loujing.com/cns-sod/

An inverse problem in distributed activation energy model is the problem of estimating unknown activation energy distribution of a thermoactive material from the measured volatile fractions emitted under heating. In this work, a closed-form approximation to the problem is derived by establishing a mathematical analogy with the inverse Fermi system problem in physics. The new solution allows for deriving distribution function f(E) from a single set of non-isothermal linear, parabolic or exponential heating data with a constant k0, compared with the existing iso-conversional integral methods in which at least three sets of linear heating data are needed. An error analysis and comparison of the proposed method along with the Miura’s is performed through case studies covering a wide range of model parameters, showing that our method is accurate enough and applicable in thermal analysis. Besides, physical implications drawn from present exploration are discussed.

Road detection is a basic task in automated driving field. In existing methods, detecting lane marks is a frequently used approach. However, road marks could fade with time, or even do not exist on some roads. Considering these factors, road region segmentation from background is much more reliable. One of the difficulties in this area is dealing with variant illumination conditions. Existing methods make some progress on reducing the influence of shadows, but they are still not satisfactory. Taking this into account, we put forward an improved shadow-free road detecting method based on color names, whose performance exceeds the existing methods. In addition, to further improving our performance on application, we adopt vanishing point detection and fuse two confidence maps to reduce the interference of sidewalk regions. Experiments on KITTI dataset depict that the proposed method is efficient. Moreover, the improved road detection method has low-complexity, which meets the requirement of practical usage.

CaC2 production via oxygen-thermal heating by combustion of coals or chars presents advantages over the conventional electro-thermal process and is increasingly gaining interest for effective use of carbon resources. In this investigation, with a particular focus on the reactor design aspect of the oxygen-thermal process, a one-dimensional unsteady pyrolysis model of bituminous coal-calcium containing composite pellets is established to describe the coupled external and internal heat transfer and decomposition rate steps. It is demonstrated by simulation that calcium hydroxide is a suitable calcium source and preferably used in a pseudo-homogeneous composite form with coal; in decomposition, the endothermic effect of calcium hydroxide is greater than that of the coal, and the endothermic effect of evaporation can be ignored; the relative importance of the involving rate steps is in the order of internal thermal resistance > external thermal resistance> reaction kinetics.

In this paper, we present an end-to-end trainable single shot text detector (SSTD) which can deal with arbitrary oriented texts in natural images. The SSTD detects texts in a single forward pass without any post-processing except for a simple non-maximum suppression. We propose a rotational prior box (RPB) mechanism to generate inclined proposals with orientation information, which enables the prior box fits the oriented text region better. The network performs prediction on multiple feature maps with different resolutions to handle text of various sizes and directly outputs bounding boxes. We further develop a dimension clustering strategy to select appropriate shapes for the default boxes that fit the text instances better. The proposed method is evaluated on three public datasets, namely ICDAR2015, MSRA-TD500 and ICDAR2013. Experiment results demonstrate its superiority in terms of effectiveness and efficiency over several state-of-the-art approaches.

The normal and generalized temperature integrals originate from the use of the Arrhenius and Harcourt–Esson equations in non-isothermal chemical kinetics analysis of solid materials, where approximate expressions to them are to be required. The aim of the present theoretical exploration is to provide both useful formulas and new insights to this problem; to the end, we address this with an extended view and new method. First, we derive an analytical connection between the two temperature integrals in terms of the Stieltjes integral. Secondly, based on this finding, a unified approach to obtaining approximate expressions of them is developed, which makes use of mathematical relationships between the Stieltjes integral, Stieltjes series and Pade approximant and leads to new Pade type approximations to the two integrals. Lastly, an error analysis and comparison of the proposed approximations along with published ones is performed, showing that they are accurate enough and applicable in thermal analysis.

GNSS (global navigation satellite system) and SINS (strap-down inertial navigation system) integrated navigation systems have been the apparatus for providing reliable and stable position and velocity information (PV). Commonly, there are two solutions to improve the GNSS/SINS integration navigation system accuracy, i.e., employing GNSS with higher position accuracy in the integration system or utilizing the high-grade inertial measurement unit (IMU) to construct the integration system. However, technologies such as RTK (real-time kinematic) and PPP (precise point positioning) that improve GNSS positioning accuracy have higher costs and they cannot work under high dynamic environments. Also, an IMU with high accuracy will lead to a higher cost and larger volume, therefore, a low-cost method to enhance the GNSS/SINS integration accuracy is of great significance. In this paper, multiple receivers based on the GNSS/SINS integrated navigation system are proposed with the aim of providing more precise PV information. Since the chip-scale receivers are cheap, the deployment of multiple receivers in the GNSS/SINS integration will not significantly increase the cost. In addition, two different filtering methods with central and cascaded structure are employed to process the multiple receivers and SINS integration. In the centralized integration filter method, measurements from multiple receivers are directly processed to estimate the SINS errors state vectors. However, the computation load increases heavily due to the rising dimension of the measurement vector. Therefore, a cascaded integration filter structure is also employed to distribute the processing of the multiple receiver and SINS integration. In the cascaded processing method, each receiver is regarded as an individual &ldquo, sensor&rdquo, and a standard federated Kalman filter (FKF) is implemented to obtain an optimal estimation of the navigation solutions. In this paper, a simulation and a field tests are carried out to assess the influence of the number of receivers on the PV accuracy. A detailed analysis of these position and velocity results is presented and the improvements in the PV accuracy demonstrate the effectiveness of the proposed method.

The location service provided by the Global Navigation Satellite System (GNSS) has gained great impact on our life. However, the signal blockage, jamming and spoofing attack lead to a serious impediment to its further application. The receivers are able to be aware of the signal blockage and jamming through signal processing technology. The goal of the signal blockage and jamming is to prevent the receiver from receiving the broadcast signals and make the receivers fail to provide navigation information. Unlike the signal blockage and jamming, the spoofing attack feeds the receivers false signals and make the receivers to output erroneous navigation information. Thus the spoofing attack detection is more difficult. Under the circumstance that no spoofing attack existing, each antenna receives unique signals. Consequently, the baseline between each pair of receivers can be calculated. Assuming the baseline length is subject to Gaussian distribution, then mean value is fixed with some noise between the pairs of receivers. When the spoofing attack occurs, the antennas receive the identical spoofing signals, which result the position solutions are almost the same. Under this condition, baseline length mean value becomes to be zero. The separation is employed as an indicator to detect the spoofing attack occurrence in the hypothesis test. In this paper, we analyzed the detectors based the baseline data. Three cases are investigated with different setups: (1) single fixed baseline; (2) two fixed and independent baselines; (3) Max/Min model for two independent baselines. In addition, the influence of the baseline length on the detection performance is analyzed, and the performance of three studied cases are evaluated by comparing the receiver operating characteristic (ROC) curves. The results demonstrate effectiveness of the model.

In this paper we study the Visual Odometry (VO) in the structured environment. Harris, SURF, ORB, such feature extraction methods are often employed by current VO methods. They are good at corner detection, but, tend to fail in an environment full of structure lines. Here we focus on just one direction of the gradient, and modify the SURF method to make it fit in the specified scene, namely, the Incomplete SURF method. As it is supposed to run in a VO framework, the calculation of the transition matrix between frames has to be adjusted accordingly. We tried the proposed combined SURF and Incomplete SURF method in our self-established dataset (seldom corners, a lot of structure lines), and succeeded in estimating the transition matrix where pure SURF failed.

In this paper, a bottom-up and data-driven model is introduced to detect co-salient objects from an image pair. Inspired by the biologically-plausible across-scale architecture, we propose a multi-layer fusion algorithm to extract conspicuous parts from an input image. At each layer, two existing saliency models are first combined to obtain an initial saliency map, which simultaneously codes for the color names based surrounded cue and the background measure based boundary connectivity. Then a global color cue with respect to color names is invoked to refine and fuse single-layer saliency results. Finally, we exploit the color names based distance metric to measure the color consistency between a pair of saliency maps and remove those non-co-salient regions. The proposed model can generate both saliency and co-saliency maps. Experimental results show that our model performs favorably against 14 saliency models and 6 co-saliency models on the Image Pair data set.

In this paper, we present a robust text detection approach in natural images which is based on region proposal mechanism. A powerful low-level detector named saliency enhanced-MSER extended from the widely-used MSER is proposed by incorporating saliency detection methods, which ensures a high recall rate. Given a natural image, character candidates are extracted from three channels in a perception-based illumination invariant color space by saliency-enhanced MSER algorithm. A discriminative convolutional neural network (CNN) is jointly trained with multi-level information including pixel-level and character-level information as character candidate classifier. Each image patch is classified as strong text, weak text and non-text by double threshold filtering instead of conventional one-step classification, leveraging confident scores obtained via CNN. To further prune non-text regions, we develop a recursive neighborhood search algorithm to track credible texts from weak text set. Finally, characters are grouped into text lines using heuristic features such as spatial location, size, color, and stroke width. We compare our approach with several state-of-the-art methods, and experiments show that our method achieves competitive performance on public datasets ICDAR 2011 and ICDAR 2013.

A stereo vision based road free space detection method was proposed in this paper. In the method, we firstly use semi global stereo matching to get disparity map, which then is projected to V disparity. Secondly, we use a modified Hough Transformation method to detect slope line to model road surface in V disparity map, then detect the vanishing line via road surface model and eliminate road and sky area in disparity map. Thirdly, the processed disparity map was projected to get U disparity, and in which we select intersection points of road and obstacles. Finally, dynamic programming was utilized to optimize best intersection line, which then were projected back to get free space boundary. The experimental results on KITTI datasets show that, the method can correctly detect free space boundary, and get remarkable results.

In this paper, we propose deeply supervised scene text detector (DSTD), a framework that can be learned from scratch. Our proposed method mainly addresses two problems. The first one is that state-of-the-art text detectors rely heavily on the off-the-shelf pre-trained models, which leads to several limitations including inflexibility and domain mismatch. The second problem is that unlike general objects, scene text usually appear in arbitrary orientations. Text detection using horizontal bounding boxes is inaccurate. In DSTD, we propose to regress rotated rectangles directly from horizontal default boxes to deal with multi-oriented text. Furthermore, we abandon the heavy pre-trained model from the SSD framework and incorporate dense layer-wise connections, enabling the network to be learned from scratch. The proposed method is evaluated on two public datasets, namely ICDAR2013 and ICDAR2015. Experimental results demonstrate its superiority over several state-of-the-art approaches.

The principle of the model inverse controller is discussed in order to design an unmanned rotorcraft flight control system applicable to different mission tasks.A neural network compensation controller and online algorithms of its weight coefficients are proposed,and the stability of the synthesized controller is analyzed.The rotational dynamics inverse controller and translational dynamics inverse controller for the unmanned rotorcraft are deduced.The attitude inner loop controller and trajectory outer loop controller are designed.The control distribution strategy of the driving motor is determined to control the speed of coaxial rotor.A combined maneuver flight mission task is planned to imitate automatic flight motion.The adaptive flight control system is verified with simulation,and the flight control abilities of horizontal and vertical motion,hover and heading motion are demonstrated and validated.These results show that the designed adaptive flight control system has adaptability and robustness,and can realize attitude stability and trajectory following.