Xingming Wu

An adaptive trajectory tracking control of wheeled mobile robots

An adaptive trajectory tracking method based on the kinematics model is proposed in this paper to solve the non-holonomic restriction problem of trajectory tracking of the wheeled mobile robot. After analyzing the course statement of the trajectory tracking control, the idea of the artificial field is introduced to improve the effect of the trajectory tracking, which takes the posture error of the robot into account to navigate the mobile robot. With the analysis of the controllers uniformly asymptotically stability and the simulations in MATLAB, the controller is proven to be practical and effective.

Design of open-loop controller for permanent magnet spherical motor

This paper presents an open-loop control system design for a permanent magnet spherical motor (PMSM). The proposed spherical motor consists of a ball-shaped rotor with a full circle of 8 permanent magnet (PM) poles and a spherical-shell-like stator with two layers of circumferential 24 air-core coils. Based on the torque model of the PMSM, the relationship between current input and torque output can be demonstrated. We design the open-loop controller by decoupling the spinning motion control, which is based on the torque equilibrium theory, from the inclination motion control. Two types of spinning motion control models are proposed and analyzed. The open-loop controller can make the PMSM have the ability to spin continuously and tilt to certain angle. The experiment of open-loop control system based on a personal computer and current controller is designed to validate the control method at last.

Dynamic control of a 3-DOF cable-driven robot based on backstepping technique

In this paper, the research object is 3-DOF spherical joint robot driven by four cables, which imitates the motion of peoples shoulder joint. The dynamic model of the 3-DOF cable-driven robotic shoulder joint is built by using Lagrange dynamics method so as to obtain the standard form of dynamic equation. Based on backstepping technique, the control law of the system model is designed for the robot trajectory tracking control problem. The control system framework is established in the environment of MATLAB/SUMILINK, and the control algorithm is programmed by s-function. The correctness of the robot dynamic model and the effectiveness of the designed controller are validated through the simulation analysis. These lay a good theory foundation for further research on effective control and performance improvement of the cable-driven parallel robot.

Salient object detection via region contrast and graph regularization

Detection of salient objects in an image is now gaining increasing research interest in computer vision community. In this study, a novel region-contrast based saliency detection solution involving three phases is proposed. First, a color-based super-pixels segmentation approach is used to decompose the image into regions. Second, three high-level saliency measures which could effectively characterize the salient regions are evaluated and integrated in an effective manner to produce the initial saliency map. Finally, we construct a pairwise graphical model to encourage that adjacent image regions with similar features take continuous saliency values, thus producing the more perceptually consistent saliency map. We extensively evaluate the proposed method on three public benchmark datasets, and show it can produce promising results when compared to 14 state-of-the-art salient object detection approaches.摘要创新点显著目标检测是计算机视觉领域的一个高度活跃的研究方向。在本文中我们提出了一种基于区域对比度和图正则化的显著目标检测算法。首先, 我们将输入图像分割为感知上相同的超像素区域; 然后, 我们提出三种高级的显著性特征, 这些特征可以很好地表征显著目标, 使用这些特征可以取得准确的显著目标检测表现; 最后, 我们提出了一种基于图模型的显著性优化策略, 该模型通过建模空间上下文关系来优化初始显著图, 最终生成了具有高度空间一致性的显著图。

Dynamic modelling and parameter identification of a three-degree-of-freedom spherical actuator

This paper addresses both dynamic modelling and dynamic parameter identification of a permanent magnet spherical actuator, which is capable of performing three-degree-of-freedom (DOF) motion in one single joint. The dynamic model of the spherical actuator is derived from Lagrange’s equations, but the parameters, called dynamic parameters, in the model are usually uncertain. Then the dynamic model is represented in a form that is linear in these parameters. A new identification method based on the output error (OE) method and recursive least square (LS) estimation is proposed to identify the parameters. This method only requires the current measurement of the stator coils in the identification procedure, which greatly simplifies the experimental process and improves the identification accuracy. Lastly, simulation and experimental results illustrate the effectiveness of the proposed method and its robustness to external disturbances. The proposed method can be also applied to other electromagnetic driving spherical actuators.

Exploiting multiple contexts for saliency detection

Abstract. A salient object detection method by extensively modeling contextual information in both the saliency feature extraction and the saliency optimization procedure is proposed. First, a sequence of multicontext features is extracted for each segmented image region. This multicontext feature encoding effectively represents the characteristics of image regions and is further mapped to the initial saliency value estimation using a nonlinear regressor. Second, contextual information is also utilized to optimize the initial saliency map, which is realized by constructing a region-level conditional random field (CRF). As such, the quality of the initial coarse saliency maps is promoted in a more principled manner. Third, multiple CRFs, defined over different scales of segmentation, are calculated and integrated so that different ranges of contextual information could contribute to the saliency optimization. Eventually, consistent saliency maps with uniformly highlighted salient regions and clear boundaries are generated. The proposed method is extensively evaluated on three public benchmark datasets, and experimental results demonstrate that our method can produce promising performance when compared to state-of-the-art salient object detection approaches.

Supervised training and contextually guided salient object detection

Abstract Existing salient object detection research generally focused on designing diverse saliency features and integrating them heuristically. In this paper, a novel salient object detection method is proposed by employing supervised training and contextual modeling. Gradient boosting decision trees are explored to aggregate features on segmented regions using a supervised training manner. Feature representation of hierarchically segmented regions is exploited to capture salient objects at different levels so as to extract discriminative features better. A region-level pairwise conditional random field (CRF) method is constructed to further boost the accuracy of saliency estimation as well as to improve the perceptual consistency of saliency maps. Experimental results demonstrate that the proposed method could achieve state-of-the-art performance over all public datasets. The F-measure is improved by 3.9 % , 13.0 % , 4.3 % on the MSRA-B, DUT-OMRON and HKU-IS dataset respectively, and the mean absolute error (MAE) is reduced by 31.6 % , 26.4 % and 21.2 % respectively on these three datasets.

Analysis and decoupling control of a permanent magnet spherical actuator

This paper presents the analysis and decoupling control of a spherical actuator, which is capable of performing three degree-of-freedom motion in one joint. The proposed actuator consists of a rotor with multiple PM (Permanent Magnet) poles in a circle and a stator with circumferential coils in three layers. Based on this actuator design, a decoupling control approach is developed. Unlike existing control methods that each coil is responsible for both the spinning and tilting motion, the proposed control strategy specifies the function of each coil. Specifically, the spinning motion is governed by the middle layer coils with a step control approach; while the tilting motion is regulated by upper and lower coils with a computed torque control method. Experiments have been conducted on the prototype to verify the validity of the design procedure, and the experimental results demonstrate the effectiveness of the analysis and control strategy.

Salient region detection based on binocular vision

Selective visual attention is a kind of mechanism of the primate visual system for rapidly focusing on attractive objects or regions in visual environment. Numerous visual attention models have been developed and optimized over the past decades. Most of the existing models concentrate on static monocular image, but little attention has been devoted to stereo depth information which is an important aspect of human perception. A region-based binocular saliency detection approach considering depth information is proposed in this paper. The difference of left and right image is used for computing disparity map and coarse saliency map. Hue, saturation, and intensity (HSI) color space is adopted and mean-shift algorithm is used for image segmentation. This study shows that the proposed region-based saliency computational method can effectively detect salient region, and it is more suitable for real time applications such as obstacle detection and visual navigation for its simplicity.

Bounded control of a spherical actuator based on orientation measurement feedback

This paper presents a permanent magnetic spherical actuator, which is capable of performing 3-degree-of-freedom (DOF). The orientation measurement of the spherical actuator has been a challenging problem so far. An interesting orientation measurement system for the rotor is proposed in this paper. Based on the position measurements, the trajectory tracking of the rotor with output feedback is addressed. As the current input is usually limited to certain degree in the applications, which leads to the bounded control torque input, the trajectory tracking controller with bounded torque input is designed. Simulation results indicate feasibility of the proposed controller. At last the experimental system is designed to validate the proposed orientation measurement system.