Deep learning-based super-resolution (SR) reconstruction is a critical approach which is used to generate high-resolution images from corresponding low-resolution images. However, the CNN-based methods are ineffective in capturing global information, whereas the Transformer-based methods have limited ability to model long-range dependencies caused by window self-attention. Besides, it is a challenging task to recover lost high-frequency information from downsampled images. In this paper, a Hybrid Information Enhanced Network (HIEN) is proposed for MRI super-resolution task. Specifically, we propose a Spatial-Channel Hybrid Attention (SCHA) to enhance specific semantics representation by combining spatial and channel self-attention together. To recover more high-frequency components, we propose a Dynamic High-Frequency Pass Filter (DHPF) to preserve high-frequency information adaptively in pixel-wise. The results of our extensive experiments indicate that HIEN outperforms other state-of-the-art methods.

Cochlear implantation is the most effective treatment for severe deafness, which requires accurate localization of temporal bone anatomy. Preoperative CT image segmentation is an essential technique to determine the location of relevant tissues in the temporal bone. However, manual segmentation is usually time-consuming and suffers from low accuracy due to the complex and small structures of these tissues in temporal bone CT. To address this issue, we proposed a CNN-Transformer structure-based 3D multi-structured model for the automatic segmentation of fine and complex tissues such as the cochlea, facial nerve, ossicles, vestibule and semicircular canal in the temporal bone CT. Our model adopts a new Transformer deformation structure, which effectively utilizes the spatial attention mechanism to capture feature dependencies, and uses the channel attention mechanism to fuse different channel semantic representations to improve segmentation accuracy. Extensive experiments on a private temporal bone CT dataset show that our model achieves higher DSS and JSS scores, and lower HD95 and ASSD scores for all targets compared with other existing segmentation methods, demonstrating its superior performance.

In this paper, we propose an efficient Multi-Circle detector which follows the fixed search order. The method makes use of horizontal and vertical search to realize circle detection, which is named as HVCD. First, this method computes edge areas in a given image. The edge areas could be divided into some regions by means of region growing. Each of regions could be efficiently searched to achieve not only one-pixel wide edges but edge segments as well. Next, the candidate circles can be extracted from every edge segment. Finally, the circle candidates could be validated with the help of Helmholtz principle. Experimental results demonstrate that HVCD could effectively detect circles on synthetic and natural images on the one hand; on the other hand, HVCD here could solve the weakness in the process of circle Hough transform implementation and EDcircles implementation.

We proposed a lower extremity exoskeleton for power amplification that perceives intended human motion via humanexoskeleton interaction signals measured by biomedical or mechanical sensors, and estimates human gait trajectories to implement corresponding actions quickly and accurately. In this study, torque sensors mounted on the exoskeleton links are proposed for obtaining physical human-robot interaction (pHRI) torque information directly. A Kalman smoother is adopted for eliminating noise and smoothing the signal data. Simultaneously, the mapping from the pHRI torque to the human gait trajectory is defined. The mapping is derived from the real-time state of the robotic exoskeleton during movement. The walking phase is identified by the threshold approach using ground reaction force. Based on phase identification, the human gait can be estimated by applying the proposed algorithm, and then the gait is regarded as the reference input for the controller. A proportional-integral-derivative control strategy is constructed to drive the robotic exoskeleton to follow the human gait trajectory. Experiments were performed on a human subject who walked on the floor at a natural speed wearing the robotic exoskeleton. Experimental results show the effectiveness of the proposed strategy.

For the robotically-assisted minimally invasive surgery, preoperative planning is essential towards assisting surgeons to prepare the intervention and to decide the best access to the surgical site. Many recent studies in preoperative planning have focused on the pose selection of the robot and the port placement. However, as such techniques cannot evaluate the performance of the multi-arm cooperation, their applications are constrained in real practise with multi-arm surgical robots. In this paper, the surgical workspace is divided and the subspaces are assigned with different weights to reflect the internal differences within the surgical workspace. We propose three metrics to evaluate the performance of the multi-arm surgical robot: Global Isotropy Index (GII) to measure the dexterity of one single robot arm; Cooperation Capability Index (CCI) to reflect the performance of the multi-arm cooperation; Minimum Distance Index (MDI) to describe the collision avoidance of the robotic arms. We also propose a combination of Particle Swarm Optimization (PSO) and Gaussian Process (GP) to locate the port placement and robot positioning. The proposed integrated PSO-GP-based optimization strategy is implemented on a three-arm surgical robot. Two sets of experiments are carried out to validate our method. The results demonstrate that the performance optimization strategy based on PSO-GP is capable of guiding surgeons to plan an intervention with the multi-arm surgical robot.

In robot-assisted upper limb rehabilitation, detecting the intentions of hemiplegic patients is essential towards assisting the patients to actively exercise instead of driving passive motions. Many interactive channels, such as voice, EMG and EEG, have been studied to estimate the motion intentions. However, limitations of these techniques, such as high complexity, have constrained their applications in practice. In this paper, we integrate a virtual environment and a low-cost motion sensor into a novel control strategy to detect motion intentions for a rehabilitation robot. Several bimanual motion sequences are intuitively programmed by a professional therapist for subjects to repeat. The strategy uses the unaffected arm and the programmed motion sequence to estimate the motion intentions of the affected arm. We adopt this strategy in Mirror Therapy, a widely-practised therapeutic intervention method. Experiments have been conducted to validate the control strategy.

The kinematic modelling has been applied to many controllers of under-actuated manipulators. Most of these studies assume that the control process is conducted within the workspace. However, as such a kinematic model cannot describe the situations when the stable grasping is violated in the real environment, these controllers may fail unexpectedly. In this paper, we propose a combination of kinematics based Workspace Analysis (WA) and Gaussian Process Classification (GPC) to model the success rates of control actions in the theoretical workspace. We also use the Gaussian Process Regression (GPR) to model the residual between the prediction of the WA and the ground truth data. We then apply this integrated model, Gaussian Processes enhanced Workspace Analysis (GP-WA), into an optimal controller. The optimal controller is implemented on a planar under-actuated gripper with two three-phalanx fingers. Two sets of simulation experiments are carried out to validate our method. The results demonstrate that the optimal manipulation controller based on GP-WA achieves high control accuracy for manipulating a wide range of objects.

Many recent studies describe micromanipulation systems by using complex Analytic Forward Models (AFM), but such models are difficult to build and incapable of describing unmodelable factors, such as manufacturing defects. In this work, we propose the Enhanced Analytic Forward Model (EAFM), an integrated model of the AFM and the Heteroscedastic Gaussian Processes (HGP). The EAFM can compensate the shortfalls of the AFM by training the HGP on the residual of the AFM. This also allows the HGP to learn the repeatability of the micromanipulation system. Based on the EAFM, we further contribute an optimal position controller for improving the accuracy and the repeatability. This optimal EAFM controller is implemented and tested on a three degree-of-freedom micromanipulator based micromanipulation system. Two sets of real-world experiments are carried out to verify our method. The results demonstrate that the controller using EAFM can statistically achieve higher accuracy and repeatability than solely using the AFM.

A high-precision pointing mechanism for intersatellite optical communication is presented in this paper, which employs flexure hinges as the passive joints and orients its moving platform by means of six piezoelectric actuators. The proposed mechanism features submicroradian resolution and microradian repeatability within a submilliradian pointing range. A corner-filleted flexure hinge considering manufacturing tolerance with high motion accuracy and large displacement is designed based on performance analysis, and the theoretical analysis is also validated by FEA simulation and experimental testing within a 5% error margin. The parameters of the parallel mechanism are synthesized by multiobjective optimization on the basis of inverse kinematics model. Moreover, the workspace determination and the FEA analysis of the optimized mechanism are also performed. Finally, a high-precision pointing prototype is fabricated and the performance testing has been implemented, which validate the effectiveness of the proposed system.

Recent studies on underactuated manipulation usually describe the system with a Kinematic Model (KM), which is built by adding external constraints to the standard manipulation analysis method. However, such external constraints are easily violated in a real-world dexterous manipulation task which results in significant control errors. In this work, the Enhanced Kinematic Model (E-KM), an integrated model of the KM and the Sparse Online Gaussian Process (SOGP) is proposed. The E-KM can compensate the shortfalls of the KM by on-the-fly training the SOGP on the residual between the prediction of the KM and the ground truth data. Based on the E-KM, we further contribute an optimal controller for underactuated manipulations. This optimal E-KM controller is implemented and tested on the iCub, a humanoid robot with two anthropomorphic underactuated hands. Two sets of real-world experiments are carried out to verify our method. The results demonstrate that the controller using E-KM statistically can achieve higher control accuracy than using solely using the KM for a wide range of objects.

A continuum manipulator with triangular notches is proposed for the potential medical applications, which is driven in the plane by wires embedded in bilateral symmetry channels. The focus of this present research is a mechanics-based kinematic model of the proposed continuum manipulator using the Timoshenko beam theory to map the driven load to the manipulator shape. In the proposed model, the continuum manipulator is divided into several V-shape units, each of which consists of two 2-node Timoshenko beam elements. Compared with previous approaches, our proposed model discards the constant curvature approximation, in which the distributed force caused by the interface contact between the wire and the V-shape unit is also considered simultaneously. The proposed mechanics model is validated experimentally on a segment of Nitinol flexible manipulator, which illustrates the effectiveness of our model to describe the continuum manipulator shaping.

The paper proposes a real-time tracking algorithm for a moving object with mobile robot based on vision using adaptive color matching and Kalman filter. The adaptive color matching can limit the region containing moving object on vision image plane. It can adjust color matching threshold to reduce the influence of lighting variations in the scene. Kalman filter is used as our prediction module to calculate motion vectors of moving object in the robot coordinate system. A view window containing the position of moving object estimated by Kalman filter is determined on image plane to reduce the image processing area. Color matching threshold can adjust itself adaptively in view window, which is used as an updating module. Experimental results show that the algorithm can adapt to lighting variations and has good tracking precision. It can also be implemented in real time.

Recently, model-based and autocode technology has become mature and brings many advantages in automotive software development. In order to take advantage of these changes, organization must adjust development process. This paper proposes a "V+v" method, in which the process of code design and implementation are divided into two development processes (controller algorithms and software platform) in contrast to traditional "V" development process. Control engineers develop algorithm based on model, generate codes with autocoding tools and carry out tests with the same model via different kinds of simulations. Platform engineers design software development platform based on OSEK-compliant RTOS according to results of system functional analysis. During software integration & test phase, algorithm codes are "installed" into the platform and then they are linked and built together to create whole ECU software. The method has been used for diesel ECU software development in China FAW Group Corporation and the results demonstrate benefits.

A redundant robot configured with a parallel base and a series manipulator is proposed in this paper. The position and orientation of end manipulator are considered as the result of the joints' ordinal motion according to the equivalence principle, so the close analytic solution for self-motion of this redundant robot is obtained via geometrical method. The Jacobian matrices of the parallel base and the whole robot system are solved respectively. Based on this, the solution for self-motion in two classes of singular configuration is also obtained. Finally a self-motion simulation of the redundant robot is performed via software MATLAB and ADAMS to validate the algorithm

Full parallel manipulator systems with flexure hinges are widely applied in many fields demanding high precision, whose dynamic performances influence the system positioning accuracy, so it is very necessary to analyze the structural dynamics in detail. The dynamic analyses of this kind of systems are relatively complex for they belong to spatial multi-loops mechanisms. In this paper, a simplified modeling is proposed which can estimates the natural characters. This method can provide an effective generalized model to estimate lower natural frequencies of the parallel manipulators with flexure hinges needless to deduce the complicated dynamics model. Firstly, the system kinematics model is established, based on which the flexibility matrix is deduced. And then the dynamics equation can be formulated, via which the lower natural frequencies can be calculated. It is validated that the method proposes an effective and concise estimating model for generalized parallel manipulator systems' dynamics characters

In robot-assisted surgery projects researchers should be able to make fast 3D reconstruction. Usually 2D images acquired with common diagnostic equipments such as UT, CT and MRI are not enough and complete for an accurate 3D reconstruction. There are some interpolation methods for approximating non value voxels which consume large execution time. We introduce a novel algorithm based on generalized regression neural network (GRNN) which can interpolate unknown volxes fast and reliable. The GRNN interpolation is used to produce new 2D images between each two succeeding ultrasonic images. It is shown that the composition of GRNN with image distance transformation can produce higher quality 3D shapes. The results of this method are compared with other interpolation methods practically. It shows this method can decrease overall time consumption and conserve the quality on online 3D reconstruction

Although micro-motion parallel manipulators have been used widely, their special architectures make it impossible to be applied in the condition demanding relative large workspace. In this paper, a novel large workspace flexure hinge-based parallel manipulator is presented, which can attain the sub-micron scale accuracy over the cubic centimeter workspace. This manipulator system is a 6-PSS parallel mechanism, in which piezoceramic motors actuate the prismatic joints, precision linear encoders detect the actuation displacements, and as the key technology, a kind of novel wide-range flexure hinges are utilized as passive spherical joints. Because of the adoption of wide-range flexure hinges, the whole system features large workspace differing from the conventional micro-motion flexure hinge-based parallel manipulators. The kinematics analysis of the whole mechanism is performed based on the stiffness model of the flexure hinges via FEM theory. The wide-range flexure hinges move over large space during the course of self-deformation, so the kinematics model is a typical geometrical nonlinear problem. In this paper, a Newton-Raphson increment iterative scheme for the kinematics model solution procedure of the whole mechanism is presented based on Updated Lagrange Formulation. And finally, the numerical calculation results about the theoretical model are given.

Parallel-structure flexure mechanisms are increasingly designed due to their superior characteristics. This paper explores a novel six degree-of-freedom large workspace flexure parallel mechanism based on the concept of wide-range flexure hinge, which can attain sub-micron scale accuracy over cubic centimeter motion range. The geometric dimensions of the flexure hinges utilized in this mechanism as passive joints will influence the system stiffness directly and other properties indirectly such as the workspace, load-carrying capacity, and driving-load capacity etc. In this paper, the stiffness model of individual flexure hinge is established firstly, and then the stiffness of the whole flexure mechanism is modeled via assembling stiffness matrices and formulating constraint equations. Based on the system stiffness model of the whole mechanism, the stiffness atlases' analysis is presented which provides theoretical principles for designing and developing this kind of flexure parallel mechanism in further. Finally, a 6-PSS large workspace flexure parallel mechanism prototype is proposed according to the analysis results, which will be utilized in the precision positioning.

This paper reports the development of a robotic system to assist surgeons to perform bonesetting operation, such as femur or tibia fracture, in which interlocking intramedullary nail are widespread to combine bones. As for ordinary procedures, sometimes the nails are distorted, so it is difficult to lock the screw in the bone. And in such a situation, surgeons have to operate under X-ray directly. Consequently the harm to the surgeon is very serious, and the operation effect is disillusionary. A novel robot-assisted bonesetting system is being developed to solve these problems. The system can assist surgeon to reposition fracted long bone and lock the interlocking intramedullary nail. It can be executed in three modes: telesurgery, semi-autonomous surgery and autonomous surgery. The destination is autonomous surgery., i.e. once configured by surgeons, the system can fulfill the part of bone repositioning and interlocking intramedullary nail locking. In order to accomplish interlocking intramedullary nail procedure, the system consists of automatic operating table, exact fluoroscopy unit, fracture reposition robot, guiding robot, computer control system and surgeon console, as well as some function units, which are image acquisition & processing and calibration system, tele-operation system, virtual operation simulation system and human-machine interface. All mentioned above are presented in detail in this paper.