Wendong Zheng

Wendong Zheng

Auto-generated profile, based on publicly available information

Wendong Zheng

Tianjin University of Technology

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 April 2020 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
WZWendong Zheng
Wendong Zheng
HLHuaping Liu
Huaping Liu
FSFuchun Sun
Fuchun Sun
The material attribute of an object's surface is critical to enable robots to perform dexterous manipulations or actively interact with their surrounding objects. Tactile sensing has shown great advantages in capturing material properties of an object's surface. However, the conventional classification method based on tactile information may not be suitable to estimate or infer material properties, particularly during interacting with unfamiliar objects in unstructured environments. Moreover, it is difficult to intuitively obtain material properties from tactile data as the tactile signals about material properties are typically dynamic time sequences. In this article, a visual-tactile cross-modal learning framework is proposed for robotic material perception. In particular, we address visual-tactile cross-modal learning in the lifelong learning setting, which is beneficial to incrementally improve the ability of robotic cross-modal material perception. To this end, we proposed a novel lifelong cross-modal learning model. Experimental results on the three publicly available data sets demonstrate the effectiveness of the proposed method.

9 October 2019 in IEEE Transactions on Automation Science and Engineering
Verified
Journal Article
Closed access
WZWendong Zheng
Wendong Zheng
HLHuaping Liu
Huaping Liu
BWBowen Wang
Bowen Wang
FSFuchun Sun
Fuchun Sun
To more actively perform fine manipulation tasks in the real world, intelligent robots should be able to understand and communicate the physical attributes of the material during interaction with an object. Tactile and vision are two important sensing modalities in robotic perception system. In this article, we propose a cross-modal material perception framework for recognizing novel objects. Concretely, it first adopts an object-agnostic method to associate information from tactile and visual modalities. It then recognizes a novel object by using its tactile signal to retrieve perceptually similar surface material images through the learned cross-modal correlation. This problem exhibits a challenge because data from visual and tactile modalities are highly heterogeneous and weakly paired. Moreover, the framework should not only consider cross-modal pairwise relevance but also be discriminative and generalized for unseen objects. To this end, we propose a weakly paired cross-modal adversarial learning (WCMAL) model for the visual-tactile cross-modal retrieval, which combines the advantages of deep learning and adversarial learning. In particular, the model fully considers the weak pairing problem between the two modalities. Finally, we conduct verification experiments on a publicly available data set. The results demonstrate the effectiveness of the proposed method.

5 March 2019 in IEEE Transactions on Magnetics
Verified
Journal Article
Closed access
WZWendong Zheng
Wendong Zheng
BWBowen Wang
Bowen Wang
HLHuaping Liu
Huaping Liu
XWXiaodong Wang
Xiaodong Wang
LYLi Yongjian
Li Yongjian
CZChanggeng Zhang
Changgeng Zhang
The tactile perception of an object's surface material plays an important role in many intelligent fields including object recognition, robotic grasp, and environment exploration. Enlightened by the human active tactile perception process that originates from scanning a finger on the surface, a novel finger-like tactile sensor for surface material recognition has been designed according to the transduction mechanism of cilia and Villari effect of the iron-gallium alloy. Based on the structural dynamic theory of Euler-Bernoulli beam, the linear constitutive equations of magetostrictive material and the Faraday law of electromagnetic induction, the output voltage model of the sensor under vibration induced from the fingertip-surface scanning has been established. Furthermore, a prototype of the sensor, on the macroscale, is fabricated. In order to verify the performance, the sensor is used to measure induced vibration signals with texture properties during sliding on the surface and classify five similar texture surfaces with small differences with a classification algorithm. The experimental results indicate that the proposed tactile sensor is effective and highly accurate for surface material classification. The sensor is characterized by high sensitivity and quick response. It has the potential of being miniaturized and integrated into the finger of a robotic hand to realize surface material recognition in real time.

10 January 2025 in IEEE/ASME Transactions on Mechatronics
Verified
Journal Article
Closed access
WZWendong Zheng
Wendong Zheng
DGDi Guo
Di Guo
WYWuqiang Yang
Wuqiang Yang
HLHuaping Liu
Huaping Liu
Tactile sensing plays a crucial role when the robot interacts with the environment. In order to detect various potential physical contacts, it is necessary to develop a large-area robotic skin deployed on the entire body to achieve distributed tactile perception. Electrical resistance tomography (ERT) technology has recently been introduced into continuous large-area tactile sensing. Although the ERT-based tactile sensor overcomes some limitations of traditional array-based tactile sensors in large-scale deployment, it still faces some challenges in practical applications. In this article, a large-area ERT-based tactile sensor using a piezoresistive sponge is proposed. Specifically, we fabricate a novel high-sensitivity piezoresistive sponge based on multiwalled carbon nanotubes and polyurethane sponge for pressure sensing. On this basis, combined with ERT technology, a flexible large-area tactile sensor is designed for distributed tactile detection. For the proposed tactile sensor, a data-driven imaging method is proposed to achieve high-quality tactile sensing. Experimental results show that the large-area tactile sensor using this method can achieve good spatial resolution, sensitivity, and positioning accuracy. In addition, in order to further evaluate the performance of the sensor in actual robot tasks, the proposed tactile sensor is integrated with a robot arm, and some tactile interaction experiments are conducted. The experimental results once again verify the effectiveness of the sensor.

16 July 2024 in IEEE Sensors Journal
Verified
Journal Article
Closed access
HZHui Zhang
Hui Zhang
LWLing Weng
Ling Weng
GLGuoheng Lin
Guoheng Lin
ZLZhuolin Li
Zhuolin Li
KMKai Meng
Kai Meng
WZWendong Zheng
Wendong Zheng
To enable robots to perform precise operations, it is essential to equip them with extensive electronic skin to enhance their perceptual capabilities for sensing external information. Mimicking the structure of human skin, we have designed a flexible tactile electronic skin array capable of timely responses to external tactile stimuli. The electronic skin features a sandwich structure, comprising magnetic thin film, a silicone buffer layer, and tunnel magnetoresistive elements. Experiments on point pressure, point distance, and point angle are used to understand the basic characteristics of the magnetic sensing element, and the size of the position sensing unit is determined with COMSOL simulation. The size of the array can be expanded as necessary, based on specific requirements and practical considerations. Based on the flexible tactile electronic skin array, we propose a novel position recognition method that utilizes the co-coupling of multiple magnetic sensing elements to determine the position information of contact forces. We designed a 10×10 grid experiment using CNNs to classify each grid position, and the results show a recognition accuracy of 91.2% and a resolution of 1mm.

4 March 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
DSDuanpeng Shi
Duanpeng Shi
WZWendong Zheng
Wendong Zheng
DGDi Guo
Di Guo
HLHuaping Liu
Huaping Liu
Electrical impedance tomography (EIT) is a non-invasive imaging technique, which has been widely used in the fields of industrial inspection, medical monitoring and tactile sensing. However, due to the inherent non-linearity and ill-conditioned nature of the EIT inverse problem, the reconstructed image is highly sensitive to the measured data, and random noise artifacts often appear in the reconstructed image, which greatly limits the application of EIT. To address this issue, a conditional diffusion model with voltage consistency (CDMVC) is proposed in this study. The method consists of a pre-imaging module, a conditional diffusion model for reconstruction, a forward voltage constraint network and a scheme of voltage consistency constraint during sampling process. The pre-imaging module is employed to generate the initial reconstruction. This serves as a condition for training the conditional diffusion model. Finally, based on the forward voltage constraint network, a voltage consistency constraint is implemented in the sampling phase to incorporate forward information of EIT, thereby enhancing imaging quality. A more complete dataset, including both common and complex concave shapes, is generated. The proposed method is validated using both simulation and physical experiments. Experimental results demonstrate that our method can significantly improves the quality of reconstructed images. In addition, experimental results also demonstrate that our method has good robustness and generalization performance.