Shiqiang Zhu

Improved Inverse Kinematics Algorithm Using Screw Theory for a Six-DOF Robot Manipulator

Based on screw theory, a novel improved inverse-kinematics approach for a type of six-DOF serial robot, “Qianjiang I”, is proposed in this paper. The common kinematics model of the robot is based on the Denavit-Hartenberg (D-H) notation method while its inverse kinematics has inefficient calculation and complicated solution, which cannot meet the demands of online real-time application. To solve this problem, this paper presents a new method to improve the efficiency of the inverse kinematics solution by introducing the screw theory. Unlike other methods, the proposed method only establishes two coordinates, namely the inertial coordinate and the tool coordinate; the screw motion of each link is carried out based on the inertial coordinate, ensuring definite geometric meaning. Furthermore, we adopt a new inverse kinematics algorithm, developing an improved sub-problem method along with Paden-Kahan sub-problems. This method has high efficiency and can be applied in real-time industrial operation. It is convenient to select the desired solutions directly from among multiple solutions by examining clear geometric meaning. Finally, the effectiveness and reliability performance of the new algorithm are analysed and verified in comparative experiments carried out on the six-DOF serial robot “Qianjiang I”.

An Efficient Inverse Kinematic Algorithm for a PUMA560-Structured Robot Manipulator

Abstract This paper presents an efficient inverse kinematics (IK) approach which features fast computing performance for a PUMA560-structured robot manipulator. By properties of the orthogonal matrix and block matrix, the complex IK matrix equations are transformed into eight pure algebraic equations that contain the six unknown joint angle variables, which makes the solving compact without computing the reverses of the 4×4 homogeneous transformation matrices. Moreover, the appropriate combination of related equations ensures that the solutions are free of extraneous roots in the solving process, and the wrist singularity problem of the robot is also addressed. Finally, a case study is given to show the effectiveness of the proposed algorithm.

Adaptive Robust Cascade Force Control of 1-DOF Hydraulic Exoskeleton for Human Performance Augmentation

Hydraulic exoskeleton with human–robot interaction becomes an important solution for those heavy load carrying applications. Good human motion intent inference and accurate human trajectory tracking are two challenging issues for the control of these systems, especially for hydraulically actuated exoskeleton where the nonlinear dynamics is quite complicated and various uncertainties exist. However, robust performance to model uncertainties has been ignored in most of the existing research. To regulate these control problems, an adaptive robust cascade force control strategy is proposed for 1-DOF hydraulically actuated exoskeleton, which is namely grouped into two control levels. In the high level, the integral of human–machine interaction force is minimized to generate the desired position (which can also be seen as the human motion intent). And in the low level, the accurate motion tracking of the generated human motion intent is developed. The nonlinear high-order dynamics with unknown parameters and modeling uncertainties are built, and adaptive robust control algorithms are designed in both control levels to deal with the complicated nonlinear dynamics and the effect of parametric and modeling uncertainties. Comparative simulation and experimental results indicate that the proposed approach can achieve smaller human–machine interaction force and good robust performance to various uncertainties.

Single-input adaptive fuzzy sliding mode control of the lower extremity exoskeleton based on human–robot interaction:

This article introduces a human–robot interaction controller toward the lower extremity exoskeleton whose aim is to improve the tracking performance and drive the exoskeleton to shadow the wearer with less interaction force. To acquire the motion intention of the wearer, two subsystems are designed: the first is to infer the wearer is in which phase based on floor reaction force detected by a multi-sensor system installed in the sole, and the second is to infer the motion velocity based on the multi-axis force sensor and admittance model. An improved single-input fuzzy sliding mode controller is designed, and the adaptive switching controller is combined to promote the tracking performance considering system uncertainties. Adaptation laws are designed based on the Lyapunov stability theorem. Therefore, the stability of the single-input adaptive fuzzy sliding mode control can be guaranteed. Finally, the proposed methods are applied to the lower extremity exoskeleton, especially in the swing phase. Its effectiveness is validated by comparative experiments.

Edge-preserving guided filtering based cost aggregation for stereo matching

An edge-preserving guided filter based stereo matching algorithm with adaptive support window is proposed.The combined cost measurement has good performance and robustness against radiometric differences.The edge-preserving filters based cost aggregation outperform other filter-based methods in stereo matching. Stereo matching has been widely used in various computer applications and it is still a challenging problem. In stereo matching, the filter-based stereo matching methods have achieved outstanding performance. A local stereo matching method based on adaptive edge-preserving guided filter is presented in this paper, which can achieve proper cost-volume filtering and keep edges well. We introduce a gradient vector of the enhanced image generated by the proposed filter into the cost computation and the Census transform is adopted in the cost measurement. This cost computation method is robust against radiometric variations and textureless areas. The edge-preserving guided filter approach is proposed to aggregate the cost volume, which further proves the effectiveness of edge-preserving filter for stereo matching. The experiments conducted on Middlebury benchmark and KITTI benchmark demonstrate that the proposed algorithm produces better results compared with other edge-aware filter-based methods.

Convolutional neural network based deep conditional random fields for stereo matching

A deep CRF based stereo matching algorithm with CNN is proposed.The CNN potential function learns the potentials of CRF in a CNN framework.The inference of the deep CRF model is formulated as a Recurrent Neural Network.The deep CRF based algorithm outperforms other MRF-based or CRF-based methods. Stereo matching has been studied for many years and is still a challenge problem. The Markov Random Fields (MRF) model and the Conditional Random Fields (CRF) model based methods have achieved good performance recently. Based on these pioneer works, a deep conditional random fields based stereo matching algorithm is proposed in this paper, which draws a connection between the Convolutional Neural Network (CNN) and CRF. The object knowledge is used as a soft constraint, which can effectively improve the depth estimation accuracy. Moreover, we proposed a CNN potential function that learns the potentials of CRF in a CNN framework. The inference of the CRF model is formulated as a Recurrent Neural Network (RNN). A variety of experiments have been conducted on KITTI and Middlebury benchmark. The results show that the proposed algorithm can produce state-of-the-art results and outperform other MRF-based or CRF-based methods.

Face Recognition Based on Extended Locally Linear Embedding

Face image data taken with various capturing devices are usually high dimensional and not very suitable for accurate classification. In this paper, a new face recognition method based on nonlinear dimensionality reduction is proposed. The extended locally linear embedding (ELLE) first embeds the high dimensional face data into a low dimensional hidden manifold. Then the linear discriminant analysis (LDA) is performed to find an optimal projection direction for classification. The proposed method was tested and evaluated using the AT&T and Yale face databases. Recognition rates were compared with Eigenface, Fisherface and LLE. Experimental results indicated the promising performance of the proposed method

Sliding mode control of electro-hydraulic servo system for lower-limb exoskeleton based on RBF neural network

Exoskeleton has drawn a great deal of attention recently because it can augment human strength and track humans motion. The objective of this paper is to enhance the performance of the electro-hydraulic servo system of the lower-limb exoskeleton, including model uncertainties and load disturbance. To accomplish the objective, a hybrid control method, combining sliding mode controller with RBF neural network, and disturbance observer is presented. The sliding mode controller is the main controller to control the electro-hydraulic servo system to track the desired trajectory. The RBF neural network is used to compensate the model uncertainties, and the disturbance observer is designed to deduce the external unknown load force. The simulation results show that the electro-hydraulic Servo System of the exoskeleton can achieve a better tracking performance with the proposed controller.

An improved graph-based visual localization system for indoor mobile robot using newly designed markers

To address the localization problem for a mobile robot in an indoor environment, an improved visual localization system based on artificial markers is proposed in this article. First, we will present a novel artificial marker which can be detected and identified easily and correctly. We will then introduce how it is designed, recognized, and verified. The markers are arranged on the ceiling and the mobile robot moves around on the ground to capture images with an up-facing monocular camera. The markers’ information, including the position and direction, will be obtained by processing the images. The camera’s uncertainty model is then put forward based on the distortion of the camera. The uncertainty of the obtained markers’ information is analyzed through the uncertainty model. Finally, with the obtained information, the marker map in global image coordinate system is established and optimized through a graph-based algorithm in which the edges can be updated to reduce the uncertainty using Bayes Estimation method. To verify the effectiveness of the localization system, numerous experiments have been conducted. Additionally, the proposed method has also been applied to industrial forklift to test its robustness in a factory environment.

Analytical research on energy harvesting systems for fluidic soft actuators

Energy consumption has significant influence on the working time of soft robots in mobile applications. Fluidic soft actuators usually release pressurized fluid to environment in retraction motion, resulting in dissipation of considerable energy, especially when the actuators are operated frequently. This article mainly explores the potential and approaches of harvesting the energy released from the actuators. First, the strain energy and pressurized energy stored in fluidic soft actuators are modeled based on elastic mechanics. Then, taking soft fiber-reinforced bending actuators as case study, the stored energy is calculated and its parametric characteristics are presented. Finally, two energy harvesting schematics as well as dynamic models are proposed and evaluated using numerical analysis. The results show that the control performance of the energy harvesting system becomes worse because of increased damping effect and its energy harvesting efficiency is only 14.2% due to the losses of energy conversion. The energy harvesting system in pneumatic form is a little more complex. However, its control performance is close to the original system and its energy harvesting efficiency reaches about 44.1%.