Showing papers presented at "International Conference on Mechatronics and Automation in 2017"

A novel RRT*-based algorithm for motion planning in Dynamic environments

Abstract: Sampling-based motion planning has become a powerful framework for solving complex robotic motion-planning tasks. Despite the introduction of a multitude of algorithms, most of these deal with the static case involving non-moving obstacles. In this work, we are extending our memory efficient RRT*FN algorithm to dynamic scenarios. Specifically, we retain the useful parts of the tree (the data structure storing the motion plan information) after a dynamic obstacle invalidates the solution path. We then employ two greedy heuristics to repair the solution instead of running the whole motion planning process from scratch. We call this new algorithm, RRT*FN-Dynamic (RRT*FND). To compare our method to the state-of-the-art motion planners, RRT* and RRT*FN, we conducted an extensive set of benchmark experiments in dynamic environments using two robot models: a non-holonomic mobile robot and an industrial manipulator. The results of these experiments show that RRT*FND finds the solution path in shorter time in most of the cases and verifies the efficacy of it in dynamic settings.

A landslide intelligent detection method based on CNN and RSG_R

Abstract: Geological disasters not only on the transmission line operation and maintenance are a great threat, but also occurred geological disasters to the people, brings the serious economic loss of property and state government. We propose an algorithm based on depth convolutional neural network (CNN) and an improved region growing algorithm (RSG_R) method for detection of landslide intelligence. The first visible light transmission line inspection image establish landslide detection image data set; and then the CNN of the image data sets were detected, and get the image existence landslide set; finally use rsg_r algorithm to extract the discriminant information of the image elements of disaster disaster (area, boundary and center). The experimental results verify the validity and superiority of the algorithm in two aspects of detection accuracy and sensitivity.

Green apple recognition method based on the combination of texture and shape features

Abstract: Automated harvesting requires accurate recognition of fruit in a tree canopy in uncontrolled environments. However, occlusion, variable illumination, variable appearance and texture make this task a complex challenge. Therefore, an accurate recognition algorithm needs to be studied which involves the detection of green apples within scenes of green leaves, shadow patterns, branches and other objects found in natural tree canopies. In this paper, the method is proposed which combines texture features, shape features and color features to solve the problem of segmenting the target and background of apple picking robot in complex background. The gray-scale difference statistical method is utilized to get the texture feature vector of the image. According to the texture feature vector, the support vector machine (SVM) is used to segment the image preliminar, and then the shape and color features are combined to achieve precise segmentation. Experiments show that the algorithm of this paper is superior to other algorithms in recognition rate and running speed. In addition, it has better segmentation effect for fruit with slight background occlusion.

A ZigBee position technique for indoor localization based on proximity learning

Abstract: In this paper, a new method of position along with node teaching and placement is proposed. Different from the traditional triangulation position, the proposed method employs a faster computation technique with special proximity learning and placement, which leads to reduced computational time. In position computation, the received signal strength indication (RSSI) values from ZigBee CC2431 modules (by Texas Instruments) were processed. The simple and efficiency position technique can be extended to many applications; for example, the handover mechanism for reference node, position using WSN protocols like Wi-Fi and Bluetooth, and basic indoor navigation.

Path optimization of AUV based on smooth-RRT algorithm

Abstract: Aiming at the complex working environment and high real-time requirements for Autonomous Underwater Vehicle (AUV), in order to meet the requirements of path planning, the paper proposes a routing optimization method of Smooth-RRT algorithm. The convergence and angle factor are added to improve the growth point and the exploration point of the expansion tree, so as to improve the speed and practicability of the algorithm. The greedy algorithm is used to smooth the planning path to meet the special requirements of the shortest distance and maneuverability of AUV. The simulation results show that the method can quickly complete the path search, improve the search efficiency and shorten the planning distance. The optimized path is more suitable for robot tracking and meets the requirements of AUV planning system.

SVM based simultaneous hand movements classification using sEMG signals

Abstract: Prediction of motion volitions is a practical issue in control of artificial limbs. Four classifiers are investigated in this paper to discriminate simultaneous hand movements based on pattern recognition of surface electromyographic (sEMG) signals. A sEMG signal processing tube composed of feature extraction, feature reduction and movements classification is proposed for offline myoelectric pattern recognition. Previous research was mainly devoted to individual hand movements classification. In this paper, several common tools are used for definition of movements. The results show that Support Vector Machine (SVM) outperforms the other three classifiers on both accuracy and model-training time. The user-depend classification accuracy reaches as high as 92.25% while the accuracy of user-independent is about 80%. The proposed classification method is a promising candidate to be used in prosthetic control for a rehabilitation robot in the future.

Hopf oscillator based adaptive locomotion control for a bionic quadruped robot

Abstract: This contribution aims to construct an adaptive locomotion controller for legged robots based on learning animals' rhythmic behaviors. The design of locomotion control method, different from traditional ones, is inspired from biological term of Central Pattern Generators (CPG), used to produce rhythmic motions. A symmetrically netted CPG structure consisting of coupled oscillators is presented to simply generate stable and natural gait patterns for legged robots. Moreover, gait transition among different gaits with locked phase relation is realized by introducing rotation matrix as coupling term that arbitrary phase difference can be obtained. Thus, smooth and prompt gait switching is achieved so as to perform dynamic walking. Also, in order to enhance adaptability to uneven terrains like slope, we adopt body attitude as sensory feedback to CPG network and the control signals for locomotion are modulated to accomplish dynamic walking on slope. The results from simulations and physical prototype experiment validate the feasibility of proposed control strategy.

Sequential learning for multimodal 3D human activity recognition with Long-Short Term Memory

Abstract: Capability of recognizing human activities is essential to human robot interaction for an intelligent robot. Traditional methods generally rely on hand-crafted features, which is not strong and accurate enough. In this paper, we present a feature self-learning mechanism for human activity recognition by using three-layer Long Short Term Memory (LSTM) to model long-term contextual information of temporal skeleton sequences for human activities which are represented by the trajectories of skeleton joints. Moreover, we add dropout mechanism and L2 regularization to the output of the three-layer Long Short Term Memory (LSTM) to avoid overfitting, and obtain better representation for feature modeling. Experimental results on a publicly available UTD multimodal human activity dataset demonstrate the effectiveness of the proposed recognition method.

Model predictive controller design with disturbance observer for path following of unmanned surface vessel

Abstract: A model predictive controller with disturbance observer is presented to force an automated unmanned surface vessel (USV) to follow a reference path with environment disturbance. A full-scale trials by fully instrumented USV is carried out to identify the dynamic model. The Serret-Frenet frame is used to define the tracking error, therefore the position tracking errors can be stabilized by stabilizing the state of the control model to zero. And the constrained control input of the considered system is solved by minimizing performance based on model predictive control (MPC). Simulation and experiments results are presented to validate the effectiveness of the proposed method.

Brain activity recognition with a wearable fNIRS using neural networks

Abstract: Functional Near-Infrared Spectroscopy (fNIRS) has taken the focus in the domain of Brain Computer Interfaces (BCI) in the recent years. However, there is not yet a defined standard for the treatment of data obtained through fNIRS. This study aims at evaluating the performance of deep neural networks and convolutional neural networks in the classification of different stimulus in the prefrontal cortex, caused by predefined activities: subtractions, word generation, and rest. After optimizing both types of networks, experimental results suggest deep neural networks to be more precise, but convolutional neural networks to be faster to train. It also shows that neither type of network is able to distinguish subtraction from word generation.

A design of a small mobile robot with a hybrid locomotion mechanism of wheels and multi-rotors

Abstract: We developed a small mobile robot in response to the demands in the disaster area. A hybrid locomotion mechanism of wheels and multi-rotors are proposed to realize both high locomotion performance and long-term operation. The wheels allow to highly maneuverable move in a narrow space, and the multi-rotors allow to move to a higher position. The objective of this study is to design the locomotion mechanism and develop a platform for confirming the basic locomotion performance. We attached a wheel mechanism into the assembled hobby drone and embedded an electrical system to operate the robot. The wheels also contribute to protect the multi-rotors from obstacles such as rubble. A stabilizer was proposed to stabilize the robot during running with wheels and designed to allow recover from flipping state. The significant of this work is not only improving the locomotion performance of the drone, but also increase the operating time, this leads various uses at disaster sites. In this paper, the details of the locomotion mechanism and some experimental results using the developed platform are shown.

Review analysis of voltage stability in power system

Abstract: According to the research on voltage stability of power system at home and abroad, the definition and classification of power system voltage stability are summarized and compared, and then analyzed from two aspects of static and dynamic voltage instability mechanism. In this paper, power system voltage stability analysis methods are summarized as static voltage stability analysis and dynamic voltage stability analysis, and the present research situation of the two kinds of analysis methods is analyzed and reviewed. Finally, the future development trend is prospected

Adaptive Extended Kalman Filter designing based on Non-singular Fast Terminal Sliding Mode control for robotic manipulators

Abstract: This paper presents a new method for position estimation of the robotic manipulators, based on the principles of the Extended Kalman Filter (EKF). The standard EKF suffers from performance depreciation and may even diverge from the true estimation in case the statistics of the noises which affect the system were unknown. Hence an Adaptive EKF has been proposed that has better outcome than conventional EKF in terms of robustness, convergence speed and estimation accuracy. Furthermore, the position of each joint is estimated to use in a Non-singular Fast Terminal Sliding Mode (NFTSM) controller. This controller will makes the states to reach in finite time. It also solves the singularity problem of Terminal sliding mode control. Computer simulations given for 2-DOF robot manipulator demonstrate the outperformance of the AEKF in compared with EKF. It has also been shown that the NFTSM controller has the ability to track the trajectory path properly and accurately.

Optimal UAVs formation transformation strategy based on task assignment and Particle Swarm Optimization

Abstract: A nested optimization strategy based on task assignment and Particle Swarm Optimization (PSO) is proposed to solve the optimization problem of formation transformation. Both the corresponding assigned target positions of interchangeable UAVs from initial formation to target formation and the relative position relationship between two formations are investigated in this paper. The Hungarian algorithm is used to rapidly solve the assignment problem, while the PSO is adopted to compute the optimal relative position relationship iteratively. To demonstrate the effectiveness and the universality of the proposed strategy, simulation results considered different situations are presented.

Multi-AUV cooperative task allocation based on improved contract network

Abstract: This paper studies multi-AUV cooperation as the research object, aiming at the problems that the traditional contract network has low overall efficiency, unreasonable distribution and coexistence of multiple tenders, a multi-AUV target allocation strategy based on improved contract network is proposed. This paper introduces the token ring network can improve the efficiency of AUV internal communication and solve the coexistence of multiple tenders, the introduction of task load rate indicators to solve the problem of unreasonable allocation of tasks. The combination of the two can effectively improve the overall efficiency. The simulation experiment in the three-dimensional environment shows that the improved contract network can improve the overall efficiency and make reasonable allocation scheme.

Design and fabrication of a pneumatic soft robotic gripper for delicate surgical manipulation

Abstract: Soft compliant gripping is a promising way to protect soft tissues from the grip damage caused by the high stress points in delicate surgical manipulation. In this paper, a new soft robotic gripper is designed to minimize the risk of soft tissues damage due to the over-gripping force generated by the conventional forceps. This new soft robotic gripper consists of a 3D-printed hook retractor shell, a soft inflatable actuator and two small rods. The ability of compliant grip is achieved by the inflated soft pneumatic actuator. Two small rods are used to separate the inflatable actuator into three-fingers-like bloats which can firmly grip the soft tissues by multi-contacts between the tissues and the gripper when the air pressure is applied to the pneumatic channel. In addition, it can protect the tissues against the harmful contacts with the rigid shell. The hook structure allows scooping-up motion during delicate surgical manipulation. The gripping tests and pulling force sensing experiments are carried out to evaluate the performance of the proposed soft robotic gripper.

Sliding-mode control of four wheel steering systems

Abstract: In this paper, a sliding mode control of active four-wheel steering systems is proposed in order to improve vehicle handling stability. An integral time-variant sliding surface is adopted to eliminate steady state errors, and a smooth function is used to alleviate the chattering effect. Dynamic responses of the front-wheel steering vehicles, the active four-wheel steering vehicles with LQR control and the active four-wheel steering vehicles with sliding mode control were compared. Simulation results show that the sliding mode control can track the ideal reference model and resist external disturbances.

Conceptual design of a novel magnetically actuated hybrid microrobot

Abstract: In this paper, to deal with the performance of the medical microrobot in fluid condition, we proposed a novel type of magnetically actuated hybrid microrobot. The magnetically actuated hybrid microrobot has characteristics of controllability and multi-function. It has a simple structure, a simple control strategy with a rotational magnetic field and good dynamic in fluid. The magnetically actuated hybrid microrobot is composed of microrobot body with a screw jet motion, microrobot leg with paddling motion and microrobot tail with fin motion. We designed a rotational magnetic field and an alternate magnetic field to realize the screw jet motion, paddling motion and fin motion. We carried out the evaluating experiments for screw jet motion and moving motion in a pipe. The experimental results indicated that the magnetically actuated hybrid microrobot has a good performance on flexibility.

Adaptive stair-ascending and stair-descending strategies for powered lower limb exoskeleton

Abstract: Enhancing the life quality for patients with spinal cord injuries (SCIs) by means of improving their mobility in daily life assisted by powered lower exoskeletons is practicable and significant. Being a part of typical daily activities, stairs with different sizes are supposed to be a usage scenario of exoskeletons aiming for extending patients' sphere of activities. This paper proposes strategies for automatically ascending/descending stairs with different sizes within the national standard based on the powered lower limb exoskeleton AIDER. The developed strategies employ infrared range sensors, foot pressure sensors and angle encoders to identify stairs whose sizes are computed based on kinematics. In addition, a methodology named “Stick From the Origin” aiming at recognizing stairs more accurately is proposed during the stair-ascending process. The Zero Moment Point (ZMP) stability criterion based on the foot pressures is utilized to ensure safety in the entire process. Experiments have been undertaken to testify the practicability and universality of the proposed strategies, by applying which AIDER can assist patients with various physiological characters to accomplish stair-climbing tasks.

PI speed controller design based on GA with time delay for BLDC motor using DSP

Abstract: Proportional integral (PI) based Genetic Algorithm (GA) method is applied to a brushless direct current (BLDC) motor with time delay conditions in this study. This study first defines and conceptualizes BLDC and PI controller with time delay conditions. Then describe a GA methodology and the method of combining PI controller into controlling a BLDC motor on a digital signal processor platform. The time responses of GA based PI controller design are better than those of standard PI controller design according to experimental results with different time delay conditions in this study.

Design and evaluation of a 3-degree-of-freedom upper limb rehabilitation exoskeleton robot

Abstract: As an important branch of medical robot, rehabilitaction training robot for hemiplegic upper limbs is a hot research topic. Based on motor relearning program, it covers many technology fields, such as rehabilitation medicine, human anatomy, mechanics, computer science, and robotics, etc. With the developing of auxiliary robot, there are still some disadvantages in the existent devices, such as heavy weight, few degrees of freedom, non-portable, low accuracy. These defects make it difficult for the rehabilitation robot to be applied to the actual rehabilitation practice. In this paper, we designed a rehabilitation robot with three DOFs. Compared to the previous device, this design can overcome some shortcomings which makes it flexible, portable, light weight, high accuracy and easy to wear. In this paper, the general structure of the flexible robot manipulator is introduced. The experiment is designed to test and analyse the error in its movement. The application of flexible transmission can greatly reduce the weight of the patient's arm, but may lead to errors in accuracy. Therefore, it is necessary to design experiments to test and analyse errors. The results show that the device can be easily and smartly dressed and has high accuracy.

Research on LEACH algorithm based on double cluster head cluster clustering and data fusion

Abstract: For the problems of uneven cluster distribution and excessive route energy consumption, which are caused by the clustering of low energy adaptive clustering algorithm (LEACH) in wireless sensor networks, an algorithm based on the clustering of double cluster heads and the data fusion mechanism of information entropy are proposed: In the cluster head selection algorithm, Select two levels of cluster heads in a well-divided cluster, two cluster heads perform different duties, this can be better to share the energy consumption, extend the network life cycle. Cluster head nodes use information entropy for classification and fusion, making the fusion results more accurate and data transmission more efficient. The algorithm is a good way to balance the energy consumption of nodes of a wireless sensor network composed of static location nodes, and extend the survival time of the network.

Design and implementation of complex systems using Mechatronics and Cyber-Physical Systems approaches

Abstract: This research has been leveraged in a conceptual framework which presents the relationships between Mechatronics and Cyber-Physical Systems (CPS) to get complex systems based on a concept of inter and muti-disciplinary collaboration design methodology. As it is demonstrated, a combination of tools was used, methods and techniques to conceive, design, implement and operate an UGV as a prototype using the complex system approach preparing it to a future launch with the Industry 4.0 environment.

The novel recognition method with Optimal Wavelet Packet and LSTM based Recurrent Neural Network

Abstract: In order to adaptively extract the subject-based time-frequency features of motor imagery EEG (MI-EEG) and make full use of the sequential information hidden in MI-EEG features, a Long Short-Term Memory (LSTM) based Recurrent Neural Network (RNN) is integrated with Optimal Wavelet Packet Transform (OWPT) to yield a novel recognition method, denoted as OWLR. Firstly, OWPT is applied to each channel of MI-EEG, and the improved distance criterion is used to find the optimal wavelet packet subspaces, whose coefficients are further selected as the time-frequency features of MI-EEG. Finally, a LSTM based RNN is used for classifying MI-EEG features. Experiments are conducted on a publicly available dataset, and the 5-fold cross validation experimental results show that OWLR yields relatively higher classification accuracies compared to the existing approaches. This is helpful for the future research and application of RNN in processing of MI-EEG.

Reactive power compensation in motor

Abstract: As an significant way to improve quality of electric energy, modern reactive power compensation technology has become an indispensable part in motor now. First of all, research background and importance were introduced, followed by emphatical introduces about the compensation principle and the conception of power factor in different situation, and then some of the method of motor energy saving are analyzed. At last, the optimal voltage tracking technology which is used for energy-saving is introduced briefly.

Research on impedance control based on force servo for single leg of hydraulic legged robot

Abstract: A force servo based impedance controller that allows compliant behaviours of the leg of a hydraulic legged robot has been presented in this paper. A novel velocity compensation algorithm which makes for elimination of the redundant forces is also included. Their performance is assessed on a 2-DOF hydraulic leg whose kinematics and transfer functions have been modeled. A double hydraulic cylinders experiment platform is designed for performance testing of the force tracking of the hydraulic actuators. Experimental results have shown a good performance of the impedance controller which allows arbitrary virtual stiffness and damping of the leg.

Real-time RGB-D based people detection and tracking system for mobile robots

Abstract: A real-time RGB-D based person detection and tracking system suitable is presented for mobile robots. Our approach combines RGB-D visual odometry estimation, target feature extraction, nearest point position information into a robust vision system that runs on open source robot operating system. As people detection is the most expensive component in any such integration, we invest significant effort into taking maximum advantage of the available depth information. Tracking process by European clustering method removes the noise information. When the depth information is not enough, it is reasonable to use the Cam-Shift method which is based on RGB information. Experimental results validate the feasibility and effectiveness of the proposed method.

Interaction Control for Manipulator with compliant end-effector based on hybrid position-force control

Abstract: This paper aims to explore the interaction control between manipulator and the environment. The ideal goal is enabling the manipulator to complete a series of position/force tasks and interact with the unknown environment. The attainment of this goal can be prevented that the manipulator is non-backdrivable, force can't be controlled directly, thus a compliant end-effector is adopted. Then the compliance of the manipulator and the tool is modelled based on iTaSC(instantaneous task specification using constraints), so as to relate forces and position/velocities in the output space and realize the contact force control. At last the proposed control scheme is tested on a contour tracking task to confirm that the proposed modeling approach obtains a good behavior of force interaction control.

Development of a novel wireless spiral capsule robot with modular structure

Abstract: Due to the limitations of the size of the wireless capsule robot using in diagnosis and treatment of gastrointestinal diseases, various functional modules cannot be integrated into a single capsule robot, which caused the practicality of the robot being reduced. This paper proposed the wireless spiral capsule robot with modular structure driven by the external magnetic field which have the mechanism that can combine and separate two robots have the Modular Structure. This robot system has two robots, namely the guide robot and auxiliary robot. The control system of the robot including the driven system, motion control and modular docking of the robots was presented. Both of the robots have two helical diversion grooves but the spiral directions are different between the two robots. In this way, they can move to each other under the same external magnetic field. This paper proposed a mechanism to make the auxiliary robot dock with the guide robot by the modular interface with the magnets and the modular structure is generic with multiple auxiliary robots. The 3D models of the robots were designed through the SolidWorks and fabricated by 3D printer. Then the active locomotive test for the two robots was taken using a thre-eaxis Helmholtz coils system. Finally, the feasibility of the docking action with the modular structure was verified. The developed capsule robot can be used to improve the capsule robot system and help for future medical applications.