Query：
学者姓名：吴海彬
Refining：
Year
Type
Indexed by
Source
Complex
Former Name
Co-
Language
Clean All
Abstract ：
Technology is nudging robot into automation and intelligence field, making dynamic collision avoidance indispensable in robotics. Aiming at safety control of mobile manipulator in human-robot coexistence environment, a novel artificial potential field based on danger index (DIAPF) approach is proposed. Firstly, hinged on the kinematic and dynamic model of mobile manipulators, a danger index system is established to take precautions against accident during the operation. We further improve the repulsion field and velocity repulsive field to promote dynamic obstacle avoidance in the human-robot coexistence environments. Feasibility and effectiveness of the proposed method are verified in simulation environment. © 2025 Acta Press. All rights reserved.
Keyword ：
Accident prevention Accident prevention Accidents Accidents Industrial robots Industrial robots Intelligent robots Intelligent robots Mobile robots Mobile robots Social robots Social robots
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Ye, Jinhua , Hong, Linxin , Wu, Haibin et al. AN IMPROVED APF METHOD FOR SAFETY CONTROL OF MOBILE MANIPULATOR IN HUMAN-ROBOT COEXISTENCE ENVIRONMENT [J]. | International Journal of Robotics and Automation , 2025 , 40 (3) : 184-193 . |
| MLA | Ye, Jinhua et al. "AN IMPROVED APF METHOD FOR SAFETY CONTROL OF MOBILE MANIPULATOR IN HUMAN-ROBOT COEXISTENCE ENVIRONMENT" . | International Journal of Robotics and Automation 40 . 3 (2025) : 184-193 . |
| APA | Ye, Jinhua , Hong, Linxin , Wu, Haibin , Zheng, Gengfeng . AN IMPROVED APF METHOD FOR SAFETY CONTROL OF MOBILE MANIPULATOR IN HUMAN-ROBOT COEXISTENCE ENVIRONMENT . | International Journal of Robotics and Automation , 2025 , 40 (3) , 184-193 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
PurposeThis study aims to address the issue that existing methods for limb action recognition typically assume a fixed wearing orientation of inertial sensors, which is not the case in real-world human-robot interaction due to variations in how operators wear it, installation errors, and sensor movement during operation.Design/methodology/approachTo address the resulting decrease in recognition accuracy, this paper introduced a data transformation algorithm that integrated the Euclidean norm with singular value decomposition. This algorithm effectively mitigates the impact of orientation errors on data collected by inertial sensors. To further enhance recognition accuracy, this paper proposed a method for extracting features that incorporate both time-domain and time-frequency domain features, markedly improving the algorithm's robustness. This paper used five classifiers to conduct comparative experiments on action recognition. Finally, this paper built an experimental human-robot interaction platform.FindingsThe experimental results demonstrate that the proposed method achieved an average action recognition accuracy of 96.4%, conclusively proving its effectiveness. This approach allows for the recognition of data from sensors placed in any orientation, using only training samples conducted at an orientation.Originality/valueThis study addresses the challenge of reduced accuracy in limb action recognition caused by sensor misorientation. The human-robot interaction system developed in this paper was experimentally verified to effectively and efficiently guide the industrial robot to perform tasks based on the operator's limb actions.
Keyword ：
Euclidean norm Euclidean norm Heuristic algorithm Heuristic algorithm Human-robot interaction Human-robot interaction Inertial sensor Inertial sensor Limb action recognition Limb action recognition
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Song, Chenyang , Wu, Jianxuan , Wu, Haibin . Orientation - invariant limb action recognition method for human-robot interaction [J]. | SENSOR REVIEW , 2025 , 45 (2) : 286-295 . |
| MLA | Song, Chenyang et al. "Orientation - invariant limb action recognition method for human-robot interaction" . | SENSOR REVIEW 45 . 2 (2025) : 286-295 . |
| APA | Song, Chenyang , Wu, Jianxuan , Wu, Haibin . Orientation - invariant limb action recognition method for human-robot interaction . | SENSOR REVIEW , 2025 , 45 (2) , 286-295 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
针对机器人示教编程过程中使用高斯混合模型(GMM)规划运动轨迹时存在的高斯分布个数难以选择、复现轨迹精度较低等问题,提出了一种复合的机器人运动轨迹学习策略.该策略包含动态时间规整(DTW)算法、高斯混合模型与道格拉斯-普克(DP)算法.首先,针对示教过程中采集的多条轨迹在时间长度上存在差异的问题,采用DTW 算法来统一示教轨迹在时域上的变化.其次,使用 GMM 算法对示教轨迹的特征进行提取,并利用高斯混合回归(GMR)算法将其重构为复现轨迹.在这个过程中采用DP算法来预估GMM算法的关键参数高斯分布的数量,与传统方法相比,能够简单直观地得到相对准确的参数值.利用 DP 算法对复现轨迹的数据点进行稀疏化并优化,不仅确保了机器人最终运动轨迹的精度,而且大幅减少了最终轨迹数据点的数量.最后,进行了不同形状的模拟焊接轨迹学习规划实验.结果表明:经由 DTW 对齐后的示教轨迹具有更加明显的运动特征,经过 GMM-GMR 学习输出的复现轨迹具有良好的表征结果;在使用GMM-GMR算法学习示教轨迹的过程中,采用DP算法可以有效预估高斯分布个数;经过DP算法稀疏化并优化的最终轨迹的平均位置误差均在 0.500 mm以内,其最大误差可以控制在0.800 mm以内,可以满足焊接轨迹规划的精度要求,验证了该策略的有效性和优越性.
Keyword ：
动态时间规整 动态时间规整 工业机器人 工业机器人 示教编程 示教编程 轨迹复现 轨迹复现 道格拉斯-普克算法 道格拉斯-普克算法 高斯混合模型 高斯混合模型
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 肖洒 , 陈旭阳 , 叶锦华 et al. 一种基于DTW-DP-GMM的工业机器人轨迹学习策略 [J]. | 天津大学学报（自然科学与工程技术版） , 2025 , 58 (1) : 68-80 . |
| MLA | 肖洒 et al. "一种基于DTW-DP-GMM的工业机器人轨迹学习策略" . | 天津大学学报（自然科学与工程技术版） 58 . 1 (2025) : 68-80 . |
| APA | 肖洒 , 陈旭阳 , 叶锦华 , 吴海彬 . 一种基于DTW-DP-GMM的工业机器人轨迹学习策略 . | 天津大学学报（自然科学与工程技术版） , 2025 , 58 (1) , 68-80 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
针对随机采样一致性(random sample consensus,RANSAC)算法对含有噪声的点云数据进行平面拟合时效果不佳和容易产生误识别的问题,对算法进行改进.通过基于密度的噪声应用空间聚类(density-based spatial clustering of applications with noise,DBSCAN)算法改变RANSAC算法初始点集合的选择策略,并使用主成分分析法(principal component analysis,PCA)计算点云各点法向量,以点到平面距离以及点的法向量与平面法向量夹角两个约束条件同时作为RANSAC算法平面拟合模型内点判定的准则.采用无噪声与分别含有300个噪声点和500个噪声点的点云仿真数据进行测试,本文算法拟合结果均接近理论值且内点距离标准差分别为1.007×10-8、0.003、0.007,优于RANSAC算法.采用实际工件点云数据在两种工况场景下进行测试,本文算法拟合平面内点比率相对于传统RANSAC算法分别提高24.7%和24.6%,平面提取完整度及准确率同样优于RANSAC算法.仿真模拟及实例分析证明了本文算法的有效性.
Keyword ：
主成分分析 主成分分析 噪声 噪声 密度聚类 密度聚类 点云平面拟合 点云平面拟合 随机采样一致性 随机采样一致性
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 叶锦华 , 林旭敏 , 吴海彬 . 基于DBSCAN的改进RANSAC点云平面拟合算法 [J]. | 湖南大学学报（自然科学版） , 2025 , 52 (2) : 76-87 . |
| MLA | 叶锦华 et al. "基于DBSCAN的改进RANSAC点云平面拟合算法" . | 湖南大学学报（自然科学版） 52 . 2 (2025) : 76-87 . |
| APA | 叶锦华 , 林旭敏 , 吴海彬 . 基于DBSCAN的改进RANSAC点云平面拟合算法 . | 湖南大学学报（自然科学版） , 2025 , 52 (2) , 76-87 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
PurposeImitation learning is a powerful tool for planning the trajectory of robotic end-effectors in Cartesian space. Present methods can adapt the trajectory to the obstacle; however, the solutions may not always satisfy users, whereas it is hard for a nonexpert user to teach the robot to avoid obstacles in time as he/she wishes through demonstrations. This paper aims to address the above problem by proposing an approach that combines human supervision with the kernelized movement primitives (KMP) model.Design/methodology/approachThis approach first extracts the reference database used to train KMP from demonstrations by using Gaussian mixture model and Gaussian mixture regression. Subsequently, KMP is used to modulate the trajectory of robotic end-effectors in real time based on feedback from its interaction with humans to avoid obstacles, which benefits from a novel reference database update strategy. The user can test different obstacle avoidance trajectories in the current task until a satisfactory solution is found.FindingsExperiments performed with the KUKA cobot for obstacle avoidance show that this approach can adapt the trajectories of the robotic end-effector to the user's wishes in real time, including trajectories that the robot has already passed and has not yet passed. Simulation comparisons also show that it exhibits better performance than KMP with the original reference database update strategy.Originality/valueAn interactive learning approach based on KMP is proposed and verified, which not only enables users to plan the trajectory of robotic end-effectors for obstacle avoidance more conveniently and efficiently but also provides an effective idea for accomplishing interactive learning tasks under constraints.
Keyword ：
Kernelized movement primitives (KMP) Kernelized movement primitives (KMP) Obstacle avoidance Obstacle avoidance Physical human-robot interaction Physical human-robot interaction Trajectory adaptation Trajectory adaptation
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Xiao, Sa , Chen, Xuyang , Lu, Yuankai et al. A KMP-based interactive learning approach for robot trajectory adaptation with obstacle avoidance [J]. | INDUSTRIAL ROBOT-THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH AND APPLICATION , 2024 , 51 (2) : 326-339 . |
| MLA | Xiao, Sa et al. "A KMP-based interactive learning approach for robot trajectory adaptation with obstacle avoidance" . | INDUSTRIAL ROBOT-THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH AND APPLICATION 51 . 2 (2024) : 326-339 . |
| APA | Xiao, Sa , Chen, Xuyang , Lu, Yuankai , Ye, Jinhua , Wu, Haibin . A KMP-based interactive learning approach for robot trajectory adaptation with obstacle avoidance . | INDUSTRIAL ROBOT-THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH AND APPLICATION , 2024 , 51 (2) , 326-339 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
工业机器人在制造业已得到广泛应用,快速、准确、自动获得目标运动点位和轨迹可以大幅提升工业机器人编程与作业效率,并可应对变化的作业环境.以自动化制造常见的平面类工件角接结合缝自动识别与跟踪为应用场景,提出基于3D点云的工件特征识别与工业机器人运动跟踪系统.通过结构光相机获得3D点云,对点云数据进行预处理、滤波、工件平面拟合、平面交叉获得结合缝点云,以及对结合缝点云进行必要的插值和顺滑处理,最终获得结合缝空间分布轨迹.再把该轨迹信息经过坐标变换和姿态补齐后,提供给工业机器人实现轨迹跟踪.实验结果表明,系统可以实现结合缝作为工件特征的自动识别与工业机器人自动运动跟踪功能.结合缝识别的路径点与实际结合缝轨迹平均误差不超过1.62 mm.目标特征的自动识别与跟踪可以大幅提升工业机器人适应多品种小批量生产的能力.
Keyword ：
工业机器人 工业机器人 点云数据处理 点云数据处理 特征识别 特征识别 轨迹规划 轨迹规划
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 王凤伟 , 虞增益 , 薛坤斌 et al. 工件3D点云特征识别与工业机器人运动跟踪 [J]. | 制造业自动化 , 2024 , 46 (10) : 205-212 . |
| MLA | 王凤伟 et al. "工件3D点云特征识别与工业机器人运动跟踪" . | 制造业自动化 46 . 10 (2024) : 205-212 . |
| APA | 王凤伟 , 虞增益 , 薛坤斌 , 吴海彬 . 工件3D点云特征识别与工业机器人运动跟踪 . | 制造业自动化 , 2024 , 46 (10) , 205-212 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
In the process of grasping workpieces, industrial robots often face the contradiction of excessive gripping force causing damage to the workpiece, and too little gripping force leading to slippage. To address this issue, a rapid sliding detection method was proposed using polyvinylidene fluoride ( PVDF) piezoelectric sensors as tactile perception elements. First, the sensor signal was decomposed and reconstructed using the variational mode decomposition (VMD) optimized by the Archimedes optimization algorithm ( AOA) to reduce noise interference. Next, the time-frequency domain features of the signal were extracted to construct the signal feature set. Finally, the dung beetle optimization (DBO) algorithm was used to optimize the selection of parameters for long short-term memory networks (LSTM). The optimized parameters obtained from DBO along with the signal feature set were applied to construct the sliding detection recognition model. The proposed sliding detection method was applied to an experiment involving electric gripper grasping. Results demonstrate precise and rapid recognition of contact status, achieving 100% accuracy with recognition times under 20 ms. Based on the recognition results, the gripping force of the electric gripper can be adjusted in realtime. © 2024 Chinese Vibration Engineering Society. All rights reserved.
Keyword ：
Feature Selection Feature Selection Frequency domain analysis Frequency domain analysis Grippers Grippers Image analysis Image analysis Image segmentation Image segmentation Industrial robots Industrial robots Linear programming Linear programming Piezoelectric transducers Piezoelectric transducers Tactile sensors Tactile sensors Time domain analysis Time domain analysis Variational mode decomposition Variational mode decomposition Variational techniques Variational techniques
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wu, Haibin , Huang, Liwen . Rapid detection of contact sliding using PVDF piezoelectric sensors [J]. | Journal of Vibration and Shock , 2024 , 43 (24) : 135-144 . |
| MLA | Wu, Haibin et al. "Rapid detection of contact sliding using PVDF piezoelectric sensors" . | Journal of Vibration and Shock 43 . 24 (2024) : 135-144 . |
| APA | Wu, Haibin , Huang, Liwen . Rapid detection of contact sliding using PVDF piezoelectric sensors . | Journal of Vibration and Shock , 2024 , 43 (24) , 135-144 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
PurposePolishing is a crucial process in mechanical manufacturing. The use of industrial robots to automate polishing is an inevitable trend in future developments. However, current robotic polishing tools are too large to reach inside deep holes or grooves in workpieces. This study aims to use a pneumatic artificial muscle (PAM) as the actuator and designs a force-controlled end-effector to reach inside the deep and narrow areas in the workpiece.Design/methodology/approachThis approach first addresses the challenge of converting the tensile force generated by the PAM into pushing force through mechanism design. In addition, a dynamics model of the end-effector was established based on the three-element model of the PAM. A combined control strategy was proposed to enhance force control accuracy and adaptability during the polishing process.FindingsExperiments were conducted on a robotic platform equipped with the proposed end-effector. The experimental results demonstrate that the end-effector can polish the inner end face of holes or grooves with diameters as small as 80 mm and depths reaching 200 mm. By implementing the combined control strategies, the target force tracking error was reduced by 48.66% compared to the use of the PID controller alone.Originality/valueA new force-controlled end-effector based on the PAM is designed for robotic polishing. According to the experimental result, this end-effector can polish not only the outer surfaces of the workpiece but also the internal surfaces of workpieces with deep holes or grooves specifically. By using the combined control strategy proposed in this paper, the end-effector significantly improves force control precision and polishing quality.
Keyword ：
Deep holes or grooves polishing Deep holes or grooves polishing Force-controlled end-effector Force-controlled end-effector Industrial robot Industrial robot Pneumatic artificial muscle Pneumatic artificial muscle
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wu, Jianxuan , Song, Chenyang , Xiao, Sa et al. Design of a force-controlled end-effector based on the pneumatic artificial muscle for robotic polishing [J]. | INDUSTRIAL ROBOT-THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH AND APPLICATION , 2024 , 52 (3) : 403-414 . |
| MLA | Wu, Jianxuan et al. "Design of a force-controlled end-effector based on the pneumatic artificial muscle for robotic polishing" . | INDUSTRIAL ROBOT-THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH AND APPLICATION 52 . 3 (2024) : 403-414 . |
| APA | Wu, Jianxuan , Song, Chenyang , Xiao, Sa , Lu, Yuankai , Wu, Haibin . Design of a force-controlled end-effector based on the pneumatic artificial muscle for robotic polishing . | INDUSTRIAL ROBOT-THE INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH AND APPLICATION , 2024 , 52 (3) , 403-414 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
工业机器人在进行工件抓取过程中,往往存在夹持力过大使工件破损、夹持力过小导致工件滑落的矛盾.为此,提出一种接触滑动的快速检测方法,采用聚偏二氟乙烯(polyvinylidene fluoride,PVDF)压电传感器作为滑觉感知元件.首先,利用阿基米德优化算法(Archimedes optimization algorithm,AOA)优化变分模态分解(variational mode decomposition,VMD)对传感器信号进行分解与重构,降低噪声干扰;然后,提取信号的时频域特征,构建信号特征集;最后,使用蜣螂优化算法(dung beetle optimization,DBO)优化选取长短期记忆网络(long short-term memory networks,LSTM)参数,将DBO优化选取后的参数和信号特征集用于构建滑动检测识别模型.将所提滑动检测方法应用于电动夹爪抓取试验,结果表明,该方法实现了接触状态的精准快速识别,准确率达到100％,识别时间在20 ms以内,根据识别结果可实时调整电动夹爪夹持力大小.
Keyword ：
变分模态分解(VMD) 变分模态分解(VMD) 滑动检测 滑动检测 聚偏二氟乙烯(PVDF) 聚偏二氟乙烯(PVDF) 长短期记忆网络(LSTM) 长短期记忆网络(LSTM) 阿基米德优化算法(AOA) 阿基米德优化算法(AOA)
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 吴海彬 , 黄力文 . 利用PVDF压电传感器实现接触滑动的快速检测 [J]. | 振动与冲击 , 2024 , 43 (24) : 135-144 . |
| MLA | 吴海彬 et al. "利用PVDF压电传感器实现接触滑动的快速检测" . | 振动与冲击 43 . 24 (2024) : 135-144 . |
| APA | 吴海彬 , 黄力文 . 利用PVDF压电传感器实现接触滑动的快速检测 . | 振动与冲击 , 2024 , 43 (24) , 135-144 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
为了使机器人在执行复杂任务时能够及时避开障碍,提出了一种基于DP-KMP 的机器人避障交互式学习方法.首先构建了该方法的整体框架,采用分割泛化策略,实现对示教轨迹的快速分段学习和对分段轨迹的避障规划;针对学习阶段,提出了基于DP算法的轨迹分割策略以提高分割效率,并使用高斯混合模型策略提取各子轨迹的参考数据库;针对轨迹规划阶段,使用KMP模型完成轨迹复现与泛化,并引入基于人机交互反馈的参考数据库更新策略,提升了人机交互避障的成功率;针对该更新策略可能失效导致避障轨迹规划失败的问题,提出了两个相应的适用条件用于检验分割生成的子轨迹.最后,通过仿真验证了所述适用条件的有效性;真实实验结果表明,使用所提出的方法分割两个实验的示教轨迹分别仅用时0.084 和 0.107 s,KUKA协作机器人在执行不同搬运任务的过程中通过与用户的多次交互成功避开了所有静止和突然变化的障碍.
Keyword ：
人-机器人交互 人-机器人交互 核化运动基元 核化运动基元 模仿学习 模仿学习 轨迹分割 轨迹分割 道格拉斯-普克算法 道格拉斯-普克算法 避障 避障
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 肖洒 , 吕勇明 , 吴海彬 . 一种基于DP-KMP的机器人避障交互式学习方法 [J]. | 仪器仪表学报 , 2024 , 45 (11) : 65-78 . |
| MLA | 肖洒 et al. "一种基于DP-KMP的机器人避障交互式学习方法" . | 仪器仪表学报 45 . 11 (2024) : 65-78 . |
| APA | 肖洒 , 吕勇明 , 吴海彬 . 一种基于DP-KMP的机器人避障交互式学习方法 . | 仪器仪表学报 , 2024 , 45 (11) , 65-78 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
