Jinhua Ye
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jinhua Ye has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Haibin Wu
Fuzhou University
12 shared publications - Haomiao Wang
School of Mechanical Engineering and Automation, Fuzhou University, Fuzhou 350116, China
4 shared publications - Shuheng Huang
School of Mechanical Engineering and Automation, Fuzhou University, Fuzhou 350116, China
3 shared publications - Youzhi Zhang
School of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou, China
3 shared publications - Bingying Zheng
School of Mechanical Engineering and Automation Fuzhou University Fujian 350108 China
2 shared publications - Gengfeng Zheng
Fujian Special Equipment Inspection and Research Institute
2 shared publications
Latest Publications
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (12) , 14668-14675
Hemispheric Diffusion: A Compositional Generative Policy for Coordinated Bimanual Manipulation
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (12) , 14668-14675
Hemispheric Diffusion: A Compositional Generative Policy for Coordinated Bimanual Manipulation
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 31 (4) , 4014-4025
Tool-Action Comprehension: A Cross-Morphology Imitation Learning Framework for Bimanual Manipulation
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 31 (4) , 4014-4025
Tool-Action Comprehension: A Cross-Morphology Imitation Learning Framework for Bimanual Manipulation
- Open Access
Orifice Leak Detection in Atmospheric Vertical Cylindrical Storage Tanks Based on SVM
Leak detection in atmospheric vertical storage tanks is crucial for preventing environmental pollution, ensuring production safety, and reducing economic losses. This study investigates orifice leaks in vertical cylindrical storage tanks under atmospheric pressure using FLUENT 16.0. The simulation reveals a significant abrupt pressure change at the leak location. Based on the simulation findings, the actual acquired pressure signals during leakage are processed with wavelet threshold denoising, confirming the abrupt pressure change characteristic. Time-domain and waveform features of the denoised signals are extracted to establish a support vector machine (SVM)-based leak detection model. The performance of different kernel functions is compared, with the linear kernel achieving the highest accuracy of 96.55%.
Open AccessOrifice Leak Detection in Atmospheric Vertical Cylindrical Storage Tanks Based on SVM
Leak detection in atmospheric vertical storage tanks is crucial for preventing environmental pollution, ensuring production safety, and reducing economic losses. This study investigates orifice leaks in vertical cylindrical storage tanks under atmospheric pressure using FLUENT 16.0. The simulation reveals a significant abrupt pressure change at the leak location. Based on the simulation findings, the actual acquired pressure signals during leakage are processed with wavelet threshold denoising, confirming the abrupt pressure change characteristic. Time-domain and waveform features of the denoised signals are extracted to establish a support vector machine (SVM)-based leak detection model. The performance of different kernel functions is compared, with the linear kernel achieving the highest accuracy of 96.55%.
- Open Access
MomentumNet-CD: Real-Time Collision Detection for Industrial Robots Based on Momentum Observer with Optimized BP Neural Network
The accurate detection and identification of collision states in industrial robot environments is a critically important and challenging task. Deep learning-based methods have been widely applied to collision detection; however, these methods primarily rely on dynamic models and dynamic threshold settings, which are subject to modeling errors and threshold adjustment latency. To address this issue, we propose MomentumNet-CD, a novel collision detection method for industrial robots that leverages backpropagation (BP) neural networks. MomentumNet-CD extracts collision state features through a momentum observer and constructs an observation model using Mahalanobis distance. These features are then processed by an optimized three-layer BP neural network for accurate collision identification. The network is trained using a modified Levenberg–Marquardt algorithm by introducing regularization terms and continuous probability outputs. Furthermore, we developed a comprehensive acquisition system based on the Q8-USB data acquisition card and the QUARC 2.7 real-time control environment. The system integrates key hardware components including a MR-J2S-70A servo driver, ATI six-dimensional force/torque (F/T) sensor, and ISO-U2-P1-F8 isolation transmitter, and the corresponding software module is developed through MATLAB/Simulink R2022b, which achieves the high-frequency real-time acquisition of critical robot joint states. The experimental results show that the MomentumNet-CD method achieves an overall accuracy of 93.65% under five different speed conditions, and the detection delay is only 12.16 ms. Compared with the existing methods, the method shows obvious advantages in terms of the accuracy and response speed of collision detection.
Open AccessMomentumNet-CD: Real-Time Collision Detection for Industrial Robots Based on Momentum Observer with Optimized BP Neural Network
The accurate detection and identification of collision states in industrial robot environments is a critically important and challenging task. Deep learning-based methods have been widely applied to collision detection; however, these methods primarily rely on dynamic models and dynamic threshold settings, which are subject to modeling errors and threshold adjustment latency. To address this issue, we propose MomentumNet-CD, a novel collision detection method for industrial robots that leverages backpropagation (BP) neural networks. MomentumNet-CD extracts collision state features through a momentum observer and constructs an observation model using Mahalanobis distance. These features are then processed by an optimized three-layer BP neural network for accurate collision identification. The network is trained using a modified Levenberg–Marquardt algorithm by introducing regularization terms and continuous probability outputs. Furthermore, we developed a comprehensive acquisition system based on the Q8-USB data acquisition card and the QUARC 2.7 real-time control environment. The system integrates key hardware components including a MR-J2S-70A servo driver, ATI six-dimensional force/torque (F/T) sensor, and ISO-U2-P1-F8 isolation transmitter, and the corresponding software module is developed through MATLAB/Simulink R2022b, which achieves the high-frequency real-time acquisition of critical robot joint states. The experimental results show that the MomentumNet-CD method achieves an overall accuracy of 93.65% under five different speed conditions, and the detection delay is only 12.16 ms. Compared with the existing methods, the method shows obvious advantages in terms of the accuracy and response speed of collision detection.
- 14 September 2023
- Published by Wiley in Advanced Materials Technologies
- Vol. 8 (21)
Cone‐Shaped Electrodes for Capacitive Tactile Sensors with High Sensitivity and Broad Linearity Range
Flexible tactile sensors with high sensitivity typically suffer from a dramatically reduces pressure resolution with increasing pressure, resulting in narrow linear ranges and limited application scenarios. Herein, a capacitive tactile sensor based on cone‐shaped electrodes (CSE) that can maintain high sensitivity over a broad linearity range is proposed. The linear response comes from the novel sensing mechanism based on the change in the facing electrode area and the rational design of the conical architecture. Finite element analysis (FEA) confirms that the interfacial contact area between the elastic electrodes and the dielectric layer can respond linearly to pressure over a broad spectrum. Based on this strategy, the fabricated sensors perform a high sensitivity (0.23 kPa−1) and superior linearity (R2 = 0.999) across a wide pressure range of up to 130 kPa. The sensors demonstrate several key features, such as good repeatability, fast response speed, low detection limit, and high durability. These attributes enable the successful use of the sensors for monitoring artery pulses and providing weighting capabilities to robots, showing promising potential for applications in daily health monitoring and human‐machine interaction.
- 14 September 2023
- Published by Wiley in Advanced Materials Technologies
- Vol. 8 (21)
Cone‐Shaped Electrodes for Capacitive Tactile Sensors with High Sensitivity and Broad Linearity Range
Flexible tactile sensors with high sensitivity typically suffer from a dramatically reduces pressure resolution with increasing pressure, resulting in narrow linear ranges and limited application scenarios. Herein, a capacitive tactile sensor based on cone‐shaped electrodes (CSE) that can maintain high sensitivity over a broad linearity range is proposed. The linear response comes from the novel sensing mechanism based on the change in the facing electrode area and the rational design of the conical architecture. Finite element analysis (FEA) confirms that the interfacial contact area between the elastic electrodes and the dielectric layer can respond linearly to pressure over a broad spectrum. Based on this strategy, the fabricated sensors perform a high sensitivity (0.23 kPa−1) and superior linearity (R2 = 0.999) across a wide pressure range of up to 130 kPa. The sensors demonstrate several key features, such as good repeatability, fast response speed, low detection limit, and high durability. These attributes enable the successful use of the sensors for monitoring artery pulses and providing weighting capabilities to robots, showing promising potential for applications in daily health monitoring and human‐machine interaction.
- Open Access
Trajectory Planning of Dual-Robot Cooperative Assembly
Efficiency can be improved through the cooperation of a dual-robot during assembly. However, how to effectively plan a simple and smooth path in a dynamic environment is a prominent problem in the process of dual-robot cooperative assembly. In this paper, a method based on RRT-Connect algorithm for trajectory planning and post-processing for trajectory optimization is proposed. This method takes full advantage of the excellent solution ability of RRT-Connect algorithm in the complex environment so as to obtain the initial path successfully. Through post-processing, the problem of RRT-Connect non-convergence to target is optimized. We use two 6-DOF industrial robots to build an experimental platform and design a dual-robot cooperative assembly system. According to the given task, the system can generate the original collision-free path based on RRT-Connect algorithm. Then the original path is simplified by Floyd algorithm and smoothed by multi-segment Bezier curve. Finally, the time parameter is sequenced for all the path points based on the iterative method, and the effective trajectory is obtained. The experimental results show that the algorithm proposed in this paper can effectively plan and optimize the trajectory of dual-robot. Compared to other methods, this approach has a higher success rate and less planning time.
Open AccessTrajectory Planning of Dual-Robot Cooperative Assembly
Efficiency can be improved through the cooperation of a dual-robot during assembly. However, how to effectively plan a simple and smooth path in a dynamic environment is a prominent problem in the process of dual-robot cooperative assembly. In this paper, a method based on RRT-Connect algorithm for trajectory planning and post-processing for trajectory optimization is proposed. This method takes full advantage of the excellent solution ability of RRT-Connect algorithm in the complex environment so as to obtain the initial path successfully. Through post-processing, the problem of RRT-Connect non-convergence to target is optimized. We use two 6-DOF industrial robots to build an experimental platform and design a dual-robot cooperative assembly system. According to the given task, the system can generate the original collision-free path based on RRT-Connect algorithm. Then the original path is simplified by Floyd algorithm and smoothed by multi-segment Bezier curve. Finally, the time parameter is sequenced for all the path points based on the iterative method, and the effective trajectory is obtained. The experimental results show that the algorithm proposed in this paper can effectively plan and optimize the trajectory of dual-robot. Compared to other methods, this approach has a higher success rate and less planning time.
- Open Access
- 26 January 2022
- Published by MDPI AG in Micromachines
- Vol. 13 (2) , 185
New Flexible Tactile Sensor Based on Electrical Impedance Tomography
In order to obtain external information and ensure the security of human–computer interaction, a double sensitive layer structured tactile sensor was proposed in this paper. Based on the EIT (Electrical Impedance Tomography) method, the sensor converts the information from external collisions or contact into local conductivity changes, and realizes the detection of one or more contact points. These changes can be processed into an image containing positional and force information. The experiments were conducted on the actual sensor sample. The OpenCV toolkit was used to process the positional information of contact points. The distributional regularities of errors in positional detection were analyzed, and the accuracy of the positional detection was evaluated. The effectiveness, sensitivity, and contact area of the force detection were analyzed based on the result of the EIT calculations. Furthermore, multi-object tests of pressure were conducted. The results of the experiment indicated that the proposed sensor performed well in detecting the position and force of contact. It is suitable for human–robot interaction.
- 26 January 2022
- Published by MDPI AG in Micromachines
- Vol. 13 (2) , 185
Open AccessNew Flexible Tactile Sensor Based on Electrical Impedance Tomography
In order to obtain external information and ensure the security of human–computer interaction, a double sensitive layer structured tactile sensor was proposed in this paper. Based on the EIT (Electrical Impedance Tomography) method, the sensor converts the information from external collisions or contact into local conductivity changes, and realizes the detection of one or more contact points. These changes can be processed into an image containing positional and force information. The experiments were conducted on the actual sensor sample. The OpenCV toolkit was used to process the positional information of contact points. The distributional regularities of errors in positional detection were analyzed, and the accuracy of the positional detection was evaluated. The effectiveness, sensitivity, and contact area of the force detection were analyzed based on the result of the EIT calculations. Furthermore, multi-object tests of pressure were conducted. The results of the experiment indicated that the proposed sensor performed well in detecting the position and force of contact. It is suitable for human–robot interaction.
- Open Access
- 3 February 2020
- Published by MDPI AG in Micromachines
- Vol. 11 (2) , 162
Inconsistency Calibrating Algorithms for Large Scale Piezoresistive Electronic Skin
In the field of safety and communication of human-robot interaction (HRI), using large-scale electronic skin will be the tendency in the future. The force-sensitive piezoresistive material is the key for piezoresistive electronic skin. In this paper, a non-array large scale piezoresistive tactile sensor and its corresponding calibration methods were presented. Because of the creep inconsistency of large scale piezoresistive material, a creep tracking compensation method based on K-means clustering and fuzzy pattern recognition was proposed to improve the detection accuracy. With the compensated data, the inconsistency and nonlinearity of the sensor was calibrated. The calibration process was divided into two parts. The hierarchical clustering algorithm was utilized firstly to classify and fuse piezoresistive property of different regions over the whole sensor. Then, combining the position information, the force detection model was constructed by Back-Propagation (BP) neural network. At last, a novel flexible tactile sensor for detecting contact position and force was designed as an example and tested after being calibrated. The experimental results showed that the calibration methods proposed were effective in detecting force, and the detection accuracy was improved.
- 3 February 2020
- Published by MDPI AG in Micromachines
- Vol. 11 (2) , 162
Open AccessInconsistency Calibrating Algorithms for Large Scale Piezoresistive Electronic Skin
In the field of safety and communication of human-robot interaction (HRI), using large-scale electronic skin will be the tendency in the future. The force-sensitive piezoresistive material is the key for piezoresistive electronic skin. In this paper, a non-array large scale piezoresistive tactile sensor and its corresponding calibration methods were presented. Because of the creep inconsistency of large scale piezoresistive material, a creep tracking compensation method based on K-means clustering and fuzzy pattern recognition was proposed to improve the detection accuracy. With the compensated data, the inconsistency and nonlinearity of the sensor was calibrated. The calibration process was divided into two parts. The hierarchical clustering algorithm was utilized firstly to classify and fuse piezoresistive property of different regions over the whole sensor. Then, combining the position information, the force detection model was constructed by Back-Propagation (BP) neural network. At last, a novel flexible tactile sensor for detecting contact position and force was designed as an example and tested after being calibrated. The experimental results showed that the calibration methods proposed were effective in detecting force, and the detection accuracy was improved.
- Open Access
A Virtual Sensor for Collision Detection and Distinction with Conventional Industrial Robots
Physical contact inevitably occurs during robot interaction with outside environments. A robot should have the ability to detect and distinguish whether a physical interaction between a human and the robot is contact or collision, so as to ensure human safety and improve interaction performance. In this paper, a virtual sensor that can detect and distinguish contact and collision between humans and industrial robots is proposed. Based on the generalized momentum of the robot, two observers with low-pass and band-pass filter characteristics were designed in this virtual sensor to realize the robot collision detection. Using the different frequency distribution ranges of the lighter contact force signal and the heavier collision force signal, the filter parameters in the two observers were appropriately selected to distinguish between collisions and contacts in close interaction between humans and robots. The virtual sensor does not require acceleration information or inverse dynamics calculations. It only needs to sample the motor driving current and position information of the robot joint, and can easily be applied to conventional industrial robots. The experimental results show that the low-pass and band-pass torque observers can detect different force signals in real-time, and the proposed virtual sensor can be used for collision detection and distinction in human–robot interactions.
Open AccessA Virtual Sensor for Collision Detection and Distinction with Conventional Industrial Robots
Physical contact inevitably occurs during robot interaction with outside environments. A robot should have the ability to detect and distinguish whether a physical interaction between a human and the robot is contact or collision, so as to ensure human safety and improve interaction performance. In this paper, a virtual sensor that can detect and distinguish contact and collision between humans and industrial robots is proposed. Based on the generalized momentum of the robot, two observers with low-pass and band-pass filter characteristics were designed in this virtual sensor to realize the robot collision detection. Using the different frequency distribution ranges of the lighter contact force signal and the heavier collision force signal, the filter parameters in the two observers were appropriately selected to distinguish between collisions and contacts in close interaction between humans and robots. The virtual sensor does not require acceleration information or inverse dynamics calculations. It only needs to sample the motor driving current and position information of the robot joint, and can easily be applied to conventional industrial robots. The experimental results show that the low-pass and band-pass torque observers can detect different force signals in real-time, and the proposed virtual sensor can be used for collision detection and distinction in human–robot interactions.
- Open Access
A Piezoresistive Tactile Sensor for a Large Area Employing Neural Network
Electronic skin is an important means through which robots can obtain external information. A novel flexible tactile sensor capable of simultaneously detecting the contact position and force was proposed in this paper. The tactile sensor had a three-layer structure. The upper layer was a specially designed conductive film based on indium-tin oxide polyethylene terephthalate (ITO-PET), which could be used for detecting contact position. The intermediate layer was a piezoresistive film used as the force-sensitive element. The lower layer was made of fully conductive material such as aluminum foil and was used only for signal output. In order to solve the inconsistencies and nonlinearity of the piezoresistive properties for large areas, a Radial Basis Function (RBF) neural network was used. This includes input, hidden, and output layers. The input layer has three nodes representing position coordinates, X, Y, and resistor, R. The output layer has one node representing force, F. A sensor sample was fabricated and experiments of contact position and force detection were performed on the sample. The results showed that the principal function of the tactile sensor was feasible. The sensor sample exhibited good consistency and linearity. The tactile sensor has only five lead wires and can provide the information support necessary for safe human—computer interactions.
Open AccessA Piezoresistive Tactile Sensor for a Large Area Employing Neural Network
Electronic skin is an important means through which robots can obtain external information. A novel flexible tactile sensor capable of simultaneously detecting the contact position and force was proposed in this paper. The tactile sensor had a three-layer structure. The upper layer was a specially designed conductive film based on indium-tin oxide polyethylene terephthalate (ITO-PET), which could be used for detecting contact position. The intermediate layer was a piezoresistive film used as the force-sensitive element. The lower layer was made of fully conductive material such as aluminum foil and was used only for signal output. In order to solve the inconsistencies and nonlinearity of the piezoresistive properties for large areas, a Radial Basis Function (RBF) neural network was used. This includes input, hidden, and output layers. The input layer has three nodes representing position coordinates, X, Y, and resistor, R. The output layer has one node representing force, F. A sensor sample was fabricated and experiments of contact position and force detection were performed on the sample. The results showed that the principal function of the tactile sensor was feasible. The sensor sample exhibited good consistency and linearity. The tactile sensor has only five lead wires and can provide the information support necessary for safe human—computer interactions.