检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Zhang, Lu Liu, Zhiyong Zhu, Xiangyu Song, Zhan Yang, Xu Lei, Zhen Qiao, Hong The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100086 China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai200031 China The Center for Biometrics and Security Research National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing100190 China The Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences Hong Kong The Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Guangdong Shenzhen518055 China

To achieve accurate and robust object detection in the real-world scenario, various forms of images are incorporated, such as color, thermal, depth, etc. However, multi-modal data often suffer from the position shift problem, i.e. the image pair is not strictly aligned, making one object has different positions in different modalities. For the deep learning method, this problem makes it difficult to fuse multi-modal features and puzzles the CNN training. In this paper, we propose a general multi-modal detector named Aligned Region CNN (AR-CNN) to tackle the position shift problem. Firstly, a region feature alignment module with adjacent similarity constraint is designed to consistently predict the position shift between two modalities and adaptively align the cross-modal region features. Secondly, we propose a novel RoI jitter strategy to improve the robustness to unexpected shift patterns. Thirdly, we present a new multi-modal feature fusion method, which selects the more reliable feature and suppresses the less useful one via feature re-weighting. Additionally, by locating bounding boxes in both modalities and building their relationships, we provide novel multi-modal labelling named KAIST-Paired. Extensive experiments on 2D and 3D object detection, RGB-T and RGB-D datasets demonstrate the effectiveness and robustness of our method. Copyright © 2022, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Output Feedback control for MIMO systems with Disturbances Compensation

Output Feedback Control for MIMO systems with Disturbances C...

引用

European control Conference (ECC)

作者： D.E. Konovalov S.A. Vrazhevsky I.B. Furtat A.S. Kremlev Martin Cech Faculty of Control Systems and Robotics ITMO University St. Petersburg Russia The Laboratory ”Control of Complex System” Institute for Problems in Mechanical Engineering of the Russian Academy of Sciences (IPME RAS) St. Petersburg Russia Pilsen — NTIS - European Center of Excellence University of West Bohemia

ISBN: (数字)9783907144022

ISBN: (纸本)9781728188133

The paper deals with the novel feedback output control law for multiple-input multiple-output (MIMO) plant with mismatched bounded disturbances. The plant model is described by linear and Lipshitz nonlinear terms. The control law is based on the auxiliary loop method for disturbance estimation and modified backstepping method for control under mismatched disturbances. Differently from existing results based on backstepping method the proposed algorithm allows to reduce the accuracy in steady state, while the existing auxiliary loop based algorithm are feasible under matched disturbances. The results in the England Power System Benchmark illustrate the efficiency of the proposed scheme.

关键词： Backstepping Steady-state control systems Mathematical model Attenuation MIMO communication Estimation

来源：评论

学校读者我要写书评

暂无评论

RoboGlove: Design of a Tendon-Driven Robotic Glove with Differential Mechanisms

RoboGlove: Design of a Tendon-Driven Robotic Glove with Diff...

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Mehran Abbasi S. Ali A. Moosavian Center of Excellence in Robotics and Control Advanced Robotics & Automated Systems (ARAS) Lab K. N. Toosi University of Technology Tehran Iran

ISBN: (数字)9781728166049

ISBN: (纸本)9781728166056

Stroke and spinal cord injury (SCI) are two main causes of human upper-limb movement impairments. In this paper, the design of a soft robotic glove with tendon actuation for rehabilitation purposes of the human hand is presented. This robotic glove called KNTU RoboGlove is designed based on differential mechanisms. To help patients, the glove makes grasping objects with different geometries be possible using a single DC motor. To this end, the differential mechanisms are first presented, which are used in two parts of the system. The actuation unit actuates all fingers with a motor, while uses differential gears that can produce the flexion or extension motions without facing singularities. Finally, the functionality of RoboGlove is evaluated in performing two different tasks. First one is grasping a cylindrical object, while the second consists of grasping an irregular-shaped object. Obtained results reveal the merits of the proposed robotic glove.

关键词： Thumb Tendons Robots Ligaments Joints Electron tubes

来源：评论

学校读者我要写书评

暂无评论

Human Body Modeling for Ground Contact Force Estimation of RoboWalk

Human Body Modeling for Ground Contact Force Estimation of R...

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Farshid Absalan S. Ali A. Moosavian Center of Excellence in Robotics and Control Advanced Robotics & Automated Systems (ARAS) Lab K. N. Toosi University of Technology Tehran Iran

ISBN: (数字)9781728166049

ISBN: (纸本)9781728166056

RoboWalk is a lower limb rehabilitation robot for Human body weight assistance. To achieve a proper design of RoboWalk, there is a need to understand the dynamics of human and foot contact forces with the ground that is the focus of this paper. In fact, an important component of human-Robo Walk forward dynamics simulation is foot-ground contact modelling. Different models have been proposed to simulate the contact of the foot with the ground. In this study, the dynamic equations of human body is developed with 19 degrees of freedom, while three points of collision with the ground are considered for each foot. Next, the kinematics information obtained from the experimental data are introduced to the dynamical equations, and the joint torques are estimated by solving the inverse dynamics of the human model. The results are compared to the experimental data for model validation. However, it is assumed that the numerical value of the foot contact model parameters are not available, so the pelvis is constrained in such a way that the human model foot does not hit the ground. Therefore, these uncertain parameters does not affect the validation results. Next, a model is selected from the existing models to express the foot contact behavior with the ground, while such models should provide stable walking with selecting the appropriate parameters. To determine the shear and normal contact force, numerical values of the parameters are required while these values depend on the shoe and the ground material and surface. Then, the path traveled by the pelvis in the human dynamic model was compared with the path of the actual pelvic motion and their difference was used to form a cost function to determine the parameters. In the next step, by choosing appropriate values for the contact parameters of the foot and the springs and the damping of the toe joint, the path of the pelvis becomes closely identical to the actual pelvis motion. Finally, after the human gait stabilized, the de

关键词： Mathematical model Dynamics Foot Force Friction Computational modeling Robots

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approach For Object Tracking Of RoboEye

Deep Learning Approach For Object Tracking Of RoboEye

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Ahmad Moori Javad Khoramdel S. Ali A. Moosavian Center of Excellence in Robotics and Control Advanced Robotics and Automated Systems (ARAS) Lab. K. N. Toosi University of Technology Tehran Iran

ISBN: (数字)9781728166049

ISBN: (纸本)9781728166056

RoboEye is a spherical 3RRR parallel robot which has been developed for its high precision. It can provide high speeds, so can be used for fast tracking tasks. To this end, in this paper proper deep learning approaches are combined with classical control methods. Deep learning algorithms are employed to detect an object of interest among various ones in a monocular image, and then obtain an estimatation of the distance to the camera. So, simultaneous depth estimation, and object detection with a monocular camera for real time implementation is proposed here. For fast calculations, also to overcome manufacturing uncertainties, inverse kinematic equations are computed by a multi-layer perceptron (MLP) network based on real data. Finally, a classical PID controller can perform a fast tracking of the object.

关键词： Cameras Estimation Kinematics Mathematical model Object detection Robot vision systems

来源：评论

学校读者我要写书评

暂无评论

RobcWalk: Conceptual and Optimal Design

RobcWalk: Conceptual and Optimal Design

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Sina Ghanaat Mahdi Nabipour S. Ali A. Moosavian Center of Excellence in Robotics and Control Advanced Robotics and Automated Systems (ARAS) Lab. K. N. Toosi University of Technology Tehran Iran

ISBN: (数字)9781728166049

ISBN: (纸本)9781728166056

In this paper, the design procedure of the RoboWalk as a lower limb rehabilitation assisting robot is presented. First, a brief introduction of the robot is presented. There is a horizontal force component that is inevitably generated by the robot, and user may suffer from this. In order to overcome this problem, a dimensional optimization is performed to minimize this component, besides the required torque to be exerted by user at knee joint. To this end, a conceptual design of RoboWalk is described using the dimensions obtained by the aforementioned optimization. The detailed design is performed afterwards, and reasons for designing of each major element are discussed in detail. Finally, it is shown that this design is capable of fulfilling our goals and requirements using actuators with lower torque-speed specification.

关键词： Exoskeletons Force Legged locomotion Knee Torque Rails

来源：评论

学校读者我要写书评

暂无评论

Simulation Analysis of Human-RoboWalk Augmented Model

Simulation Analysis of Human-RoboWalk Augmented Model

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Mohamad R. Mohamadi Vahid Akbari Omid Mahdizadeh Mahdi Nabipour S. Ali A. Moosavian Center of Excellence in Robotics and Control Advanced Robotics and Automated Systems (ARAS) Lab. K. N. Toosi University of Technology Tehran Iran

ISBN: (数字)9781728166049

ISBN: (纸本)9781728166056

Exoskeletons are used in different applications including the area of physical therapy in order to facilitate the patient's exercises and as an assisting technology to assist the elderly carry out their ordinary activities. At the beginning of this paper, the RoboWalk assistive device is introduced. Before constructing the system, the performance of the device must be analyzed through different simulations. Hence, the dynamics of the human and RoboWalk are then obtained using the Newton-Euler (NE) and the Recursive Newton Euler Algorithm (RNEA). The obtained models are then augmented to the human model to estimate the RoboWalkjoint forces and torques, and those of the human model. then, RoboWalk is imported to the human model in Opensim software and the augmented model is obtained by defining some constraints and joint models. controllers are then designed for the human and Robo Walk in Opensim. In the next step, The NE and RNEA models are then compared to the models obtained from Opensim software using the same gait data. It is shown that the NE and RNEA methods match very closely and both of the models possess the same behavior as the Opensim model.

关键词： Mathematical model Robots Force Actuators Footwear Knee Dynamics

来源：评论

学校读者我要写书评

暂无评论

RoboWalk Multiple-Sensor and Multipolar Data Fusion

RoboWalk Multiple-Sensor and Multipolar Data Fusion

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Omid Mahdizadeh Vahid Akbari S. Ali A. Moosavian Center of Excellence in Robotics and Control Advanced Robotics and Automated Systems (ARAS) Lab. K. N. Toosi University of Technology. Tehran Iran

ISBN: (数字)9781728166049

ISBN: (纸本)9781728166056

RoboWalk is a two-legged rehabilitation assisting robot, which helps users by carrying a portion of the weight. In this paper, we propose an algorithm to correct the attitude and position error by means of an Adaptive Root Square Unscented Kalman Filter. To this end, the IMU and encoders are used to determine the robot kinematic configuration, but since there is no solid connection between the ground and the feet, the sensor's data may not lead to exact positioning configuration. Therefore, a new approach for estimating the position of the feet and the seat is presented. The method is based on using the standing foot IMU, and related encoders to obtain the seat position. Then, the obtained result is compared to the seat IMU and correcting the position. Next, the corrected position and the swing leg encoders are used to obtain the swing foot position, which is compared to the corresponding IMU, to modify and yield all corrected positions. Describing the proposed method, it is then implemented, and the results will be discussed.

关键词： Mathematical model Robot sensing systems Estimation Robot kinematics Foot

来源：评论

学校读者我要写书评

暂无评论

Real-world Robot Reaching Skill Learning Based on Deep Reinforcement Learning

Real-world Robot Reaching Skill Learning Based on Deep Reinf...

引用

第32届中国控制与决策会议

作者： Naijun Liu Tao Lu Yinghao Cai Rui Wang Shuo Wang State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences University of Chinese Academy of Sciences Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institutes of Advanced TechnologyChinese Academy of Sciences Center for Excellence in Brain Science and Intelligence Technology of the Chinese Academy of Sciences

Traditional programming method can achieve certain manipulation tasks with the assumption that robot environment is known and ***,with robots gradually applied in more domains,robots often encounter working scenes which are complicated,unpredictable,and *** overcome the limitation of traditional programming method,in this paper,we apply deep reinforcement learning(DRL) method to train robot agent to obtain skill *** policy trained with DRL on real-world robot is time-consuming and costly,we propose a novel and simple learning paradigm with the aim of training physical robot ***,our method train a virtual agent in an simulated environment to reach random target position from random initial ***,virtual agent trajectory sequence obtained with the trained policy,is transformed to real-world robot command with coordinate transformation to control robot performing reaching *** show that the proposed method can obtain self-adaptive reaching policy with low training cost,which is of great benefits for developing intelligent and robust robot manipulation skill system.

关键词： Robot reaching skill learning deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Robust Impedance control for Dual User Haptic Training System

Robust Impedance Control for Dual User Haptic Training Syste...

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： R. Heidari M. Motaharifar H. D. Taghirad Advanced Robotics and Automated Systems (ARAS) Industrial Control Center of Excellence (ICCE) Faculty of Electrical Engineering K.N. Toosi University of Technology Tehran Iran

ISBN: (数字)9781728166049

ISBN: (纸本)9781728166056

In this paper, an impedance controller with switching parameters for a dual-user haptic training system is introduced. The trainer and the trainee are connected through their haptic consoles, and the trainee performs the surgical procedure on the environment. The trainer can intervene in the procedure by pressing a mechanical pedal; thus, the control parameters are switched to transfer the authority over the task from the trainee to the trainer. The stability of each subsystem and the closed-loop stability of the overall system are investigated. The simulation results verify the performance of the proposed controller.

关键词： Haptic interfaces Impedance Surgery Training Mathematical model Stability analysis Force

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：