Dr. Junwu Zhu

The cooperative control problem of nonlinear multiagent systems is studied in this paper. The followers in the communication network are subject to unmodeled dynamics. A fully distributed neural-networks-based adaptive control strategy is designed to guarantee that all the followers are asymptotically synchronized to the leader, and the synchronization errors are within a prescribed level, where some global information, such as minimum and maximum singular value of graph adjacency matrix, is not necessarily to be known. Based on the Lyapunov stability theory and algebraic graph theory, the stability analysis of the resulting closed-loop system is provided. Finally, an numerical example illustrates the effectiveness and potential of the proposed new design techniques.

29 January 2020 in IEEE Access

Journal Article

Image Inpainting Based on Generative Adversarial Networks

Yi Jiang

Jiajie Xu

Baoqing Yang

Jing Xu

Junwu Zhu

Image inpainting aims to fill missing regions of a damaged image with plausibly synthesized content. Existing methods for image inpainting either fill the missing regions by borrowing information from surrounding areas or generating semantically coherent content from region context. They often produce ambiguous or semantically incoherent content when the missing region is large or with complex structures. In this paper, we present an approach for image inpainting. The completion model based on our proposed algorithm contains one generator, one global discriminator, and one local discriminator. The generator is responsible for inpainting the missing area, the global discriminator aims evaluating whether the repair result has global consistency, and the local discriminator is responsible for identifying whether the repair area is correct. The architecture of the generator is an auto-encoder. We use the skip-connection in the generator to improve the prediction power of the model. Also, we use Wasserstein GAN loss to ensure the stability of training. Experiments on CelebA dataset and LFW dataset demonstrate that our proposed model can deal with large-scale missing pixels and generate realistic completion results.

21 February 2024 in IEEE Transactions on Consumer Electronics

Journal Article

Research on UAVs Reconnaissance Task Allocation Method Based on Communication Preservation

Die Ge

The research on multi-UAV reconnaissance task allocation is of great significance for both military and civilian fields. Existing methods simplify targets as points and ignore the heterogeneity of shape and values. Motivated by this oversight, a reconnaissance strategy is proposed for each type of heterogeneous target. Based on the proposed strategy, the reconnaissance information revenue function is formalized considering the correlation between target and time. Additionally, the utility of UAVs depends on the revenue of information and flight costs. The Heterogeneous Target Reconnaissance Allocation (HTRA) scheme is designed to optimize the utility. Specifically, the resource matching matrix model was firstly used to realize initial allocation. Subsequently, the sequence of tasks is optimized by using the Improved Particle Swarm Optimization (IPSO) algorithm. Finally, the reconnaissance time is optimized by Gradient Projection (GP) algorithm to improve the utility of UAVs. Moreover, the HTRA scheme is expanded to HTRA+, and a communication preservation-based auction method is proposed to address limited communication in UAVs cooperation. The experiment results show that the proposed scheme outperforms existing methods in terms of reconnaissance utility and the rate of task completion.

20 December 2018 in Multimedia Tools and Applications

Journal Article

An auction-based rescue task allocation approach for heterogeneous multi-robot system

Jieke Shi

Zhou Yang

Junwu Zhu

30 January 2022 in International Journal of Machine Learning and Cybernetics

Journal Article

Small target deep convolution recognition algorithm based on improved YOLOv4

4 April 2024 in Applied Soft Computing

Journal Article

Social-aware graph contrastive learning for recommender systems

Jialuo Zhou

Yaling Xie

About Me

Junwu Zhu is the deputy dean of the School of Information Engineering at Yangzhou University. He was selected as one of the “Six Talent Peaks” of Jiangsu Province in 2009. In 2010, he won the Jiangsu Provincial Government Study Abroad Scholarship. In 2011, he was fully funded by the Intelligent Systems Laboratory of UWS University as a visiting scholar. In 2013 and 2015, he twice entered the IMAGO Laboratory of the University of Guelph in Canada as a visiting professor. In 2014, he was awarded the title of Young and Middle-aged Experts with Outstanding Contributions in Yangzhou City. In 2016, he was awarded the Science and Technology Contribution Award of Yangzhou University. In 2017, he was selected for the Jiangsu Province Mass Entrepreneurship Talent Plan (Vice President of Science and Technology). In 2018, he was selected for the Yangzhou Green Yang Jinfeng Plan Innovation Leading Talent. He is currently a senior member of the China Computer Society, a member of the Information System Committee of the China Computer Society, a communications member of the Chinese Information Processing Committee of the China Computer Society, deputy director of the Organizational Working Committee of the Jiangsu Provincial Computer Society, and a member of the Jiangsu Provincial Service and Network Computing Committee. He is mainly engaged in the teaching and research of knowledge engineering and algorithmic game theory in the field of artificial intelligence.

Dr. Junwu Zhu

Dr. Junwu Zhu

Research Focus & Metrics

Academic Career Length

Publication Type Breakdown

Representative Publications

About Me