BIOGRAPHY

Research Interests

Professor Junfeng WU's research interests mainly include distributed systems, control network systems, Kalman filtering, and signal processing, cyber security and privacy.

Education Background

  • Ph.D., (Sept. 2009 – Aug. 2013)
    Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST)
    Advisor: Ling Shi.
    Thesis: Event-based State Estimation: Theory and Applications, Excellent Grade
  • Visiting Student (May 2012 – Aug. 2012)
    Department of Automatic Control, KTH, Sweden
    Advisor: Karl H. Johansson
  • B.Eng., (Sep. 2005 – Jul. 2009)
    Department of Control Science and Engineering, Zhejiang University, Hangzhou, P.R.China
    Outstanding Graduate
  • Working Experiences

  • Associate Professor (Sep. 2021 – present)
    School of Data Science, Chinese University of Hong Kong Shenzhen, Shenzhen, P. R. China
  • Professor (on tenure track) (Jul. 2017 – Sep.2021)
    College of Control Science and Engineering, Zhejiang University, Hangzhou, P. R. China
  • Visiting Scholar (Apr. 2016 – Aug. 2016)
    College of Engineering and Computer Science, The Australian National University, Australia
  • Researcher (Jan. 2016 – Jun. 2017)
    Department of Automatic Control, The Royal Institute of Technology (KTH), Sweden
  • Postdoctoral Researcher (Jan. 2014 – Dec. 2015)
    Department of Automatic Control, KTH, Sweden
    Advisor: Karl H. Johansson
  • Research Associate (Sep. 2013 – Dec. 2013)
    Department of Electronic and Computer Engineering, HKUST, Hong Kong
    Advisor: Ling Shi
  • Awards

    • Invitational Fellowships for Research in Japan, The Japan Society for the Promotion of Science (JSPS), Japan. 2019
    • Endeavour Research Fellowship, Australian Government.2016
    • Guan Zhao-Zhi Best Paper Award, The 34th Chinese Control Conference.2015
    • Outstanding Reviewer, Elsevier Automatica.2014
    • Outstanding Reviewer, IEEE Transactions on Control of Network Systems.2014

    NEWS

    • A paper was accepted by Automatica (2022.08.08)
      Biqiang Mu, Tianshi Chen, He Kong, Bo Jiang, Lei Wang and Junfeng Wu. On Embeddings and Inverse Embeddings of Input Design for Regularized System Identification. Automatica, accepted.
    • A paper was accepted by TSP (2022.07.30)
      Guangyang Zeng, Biqiang Mu, Jiming Chen, Zhiguo Shi and Junfeng Wu. Global and Asymptotically Efficient Localization from Range Measurements. IEEE Transactions on Signal Processing,accepted.
    • Congratulations to Shiyu Chen for receiving CUHK-Shenzhen research reward! (2022.07.28)
    • Welcome Guangyang Zeng(曾广扬)as a postdoctoral researcher to join us! (2022.07.11)
    • Welcome Yuan Shen(沈源), Wentao Wang(王文涛), Shiyu Chen(陈诗雨) to visit us! (2022.07.11)

    PUBLICATION

    JOURNAL PAPERS

    [43]Guangyang Zeng, Biqiang Mu, Jiming Chen, Zhiguo Shi and Junfeng Wu. "Global and Asymptotically Efficient Localization from Range Measurements", IEEE Transactions on Signal Processing, accepted.
    [42] Xinghan Li, Haodong Jiang, Xingyu Chen, He Kong, and Junfeng Wu. "Closed-form Error Propagation on $SE_n(3)$ Group for Invariant EKF with Applications to VINS", IEEE Robotics and Automation Letters, accepted.
    [41] Guangyang Zeng, Biqiang Mu, Jieqiang Wei, Wing Shing Wong, and Junfeng Wu, “Localizability with Range-Difference Measurements: Numerical Computation and Error Bound Analysis”, IEEE/ACM Transactions on Networking, Early Access, DOI: 10.1109/TNET.2022.3162930.
    [40] Biqiang Mu, Tianshi Chen, He Kong, Bo Jiang, Lei Wang,Junfeng Wu. On Embeddings and Inverse Embeddings of Input Design for Regularized System Identification. Automatica,accepted.
    [39] Xingkang He, Yu Xing, Junfeng Wu, Karl H. Johansson. Event-Triggered Distributed Estimation With Decaying Communication Rate. SIAM Journal on Control and Optimization, v.60, n.2, pp.992-1017, 2022.
    [38] Guangyang Zeng, Xiaoqiang Ren, and Junfeng Wu, “Low-complexity Distributed Detection with One-bit Memory Under Neyman-Pearson Criterion”, IEEE Transactions on Control of Network Sys-tems, 2022, Early Access. DOI: 10.1109/TCNS.2022.3141020.
    [37] Lingying Huang, Junfeng Wu, Yilin Mo, and Ling Shi, “Joint Sensor and Actuator Placement for Infinite-Horizon LQG Control”, IEEE Transactions on Automatic Control, accepted.
    [36] Yuqing Ni, Junfeng Wu, Li Li, and Ling Shi, “Multi-Party Dynamic State Estimation That Preserves Data and Model Privacy”, IEEE Transactions on Information Forensics and Security, vol. 16, pp. 2288-2299, 2021.
    [35] Kemi Ding, Junfeng Wu, Lihua Xie, ”Minimum-Degree Distributed Graph Filter Design”, IEEE Transactions on Signal Processing, vol. 69, pp. 1083-1096, 2021.
    [34] Yuchi Wu, Junfeng Wu, Mingyi Huang, and Ling Shi, “Mean-Field Transmission Power Control in Dense Networks”, IEEE Transactions on Control of Network Systems,vol. 8, pp.99-110, 2020.
    [33] Takuya Iwaki, Junfeng Wu, Yuchi Wu, Henrik Sandberg, Karl H. Johansson, “Multi-hop Sensor Network Scheduling for Optimal Remote Estimation”, Automatica, vol. 127, 109498, 2021.
    [32] Lingying Huang, Jiazheng Wang, Enoch Kung, Yilin Mo, Junfeng Wu, and Ling Shi, “Stochastic Event-based Sensor Schedules for Remote State Estimation in Cognitive Radio Sensor Networks”, IEEE Transactions on Automatic Control , vol. 66, pp. 2407-2414, 2020.
    [31] Tao Yang, Jemin George, Jiahu Qin, Xinlei Yi, and Junfeng Wu, “Distributed Finite-time Least Squares Solver for Network Linear Equations”, Automatica, vol. 113 108798, 2020.
    [30] Jiahu Qin, Weiming Fu, Junfeng Wu, Wei Xing Zheng, and Yu Kang, “Interval Consensus over Random Networks”, Automatica, vol. 111, 108603, 2020.
    [29] Enoch Kung, Jiazheng Wang, Junfeng Wu, Dawei Shi, and Ling Shi, “On the Nonexistence of Event Triggers that preserve Gaussian State in Presence of Packet-Drop”, IEEE Transactions on Automatic Control, vol. 65, pp. 4302-4307, 2019.
    [28] Yuan Huang, Junzheng Wang, Dawei Shi, Junfeng Wu, and Ling Shi, “Event-triggered Sampled-data Control: an Active Disturbance Rejection Approach”, IEEE/ASME Transactions on Mechatronics, vol. 24, pp. 2052-2063, 2019.
    [27] Tao Yang, Xinlei Yi, Junfeng Wu, Ye Yuan, Di Wu, Ziyang Meng, Yiguang Hong, Hong Wang, ZongliLin, and Karl H Johansson, “A survey of Distributed Optimization”, Annual Reviews in Control, vol. 47, pp. 278-305, 2019.
    [26] Xinlei Yi, Tao Yang, Junfeng Wu, and Karl H. Johansson, “Distributed Event-Triggered Control for Global Consensus of Multi-Agent Systems with Input Saturation”, Automatica, vol. 100, pp. 1-9, Feb., 2019, (regular paper).
    [25] Junfeng Wu, Guodong Shi, Brian D. O. Anderson, and Karl H. Johansson, “Kalman Filtering over Fading Channels: Zero-One Laws and Almost Sure Stabilities”, IEEE Transactions on Information Theory, vol. 64, pp. 6731-6742, Oct., 2018, (regular paper).
    [24] Junfeng Wu, Guodong Shi, Brian D. O. Anderson, and Karl H. Johansson, “Kalman Filtering over Gilbert-Elliott Channels: Stability Conditions and the Critical Curve”, IEEE Transactions on Automatic Control, vol. 64, pp. 1003-1017, Apr., 2018, (regular paper).
    [23] Heng Zhang, Yifei Qi, Junfeng Wu, Lingkun Fu, and Lidong He, “DOS Attack Energy Management against Remote State Estimation”, IEEE Transactions on Control of Network Systems, vol. 5, pp. 383-394, Mar. 2018.
    [22] Xiaoqiang Ren, Junfeng Wu, Karl Henrik Johansson, Guodong Shi, and Ling Shi, “Infinite Hori- zon Optimal Transmission Power Control for Remote State Estimation over Fading Channels”, IEEE Transactions on Automatic Control, vol. 63, pp. 85-100, Jan., 2018, (regular pa- per).
    [21] Xiufang Shi, and Junfeng Wu, “To Hide Private Position Information in Localization Using Time Difference of Arrival”, IEEE Transactions on Signal Processing, vol.66, issue.18, pp. 4946 - 4956, Sep., 2018, (regular paper).
    [20] Bo Li, Junfeng Wu, Hongsheng Qi, Alexandre Proutiere, and Guodong Shi, “Boolean Gossiping Networks”, IEEE Transactions on Networking, vol. 26, pp. 118-130, Feb., 2018, (regular paper).
    [19] Xiaoqiang Ren, Junfeng Wu, Subhrakantim Dey, and Ling Shi, “Attack Allocation on Remote State Estimation in Multi-Systems: Structural Results and Asymptotic Solution”, Automatica, vol. 87, pp. 184-194, Jan., 2018, (regular paper).
    [18] Yuzhe Li, Junfeng Wu, and Tongwen Chen, “Transmit Power Control and Remote State Estimationwith Sensor Networks: A Bayesian Inference Approach”, Automatica, vol. 97. pp. 292-300, Nov., 2018.
    [17] Duo Han, Junfeng Wu, Yilin Mo, and Lihua Xie, “On Stochastic Sensor Network Scheduling for Multiple Processes”, IEEE Transactions on Automatic Control, vol. 62, pp. 6633- 6640, Dec., 2017.
    [16] Junfeng Wu, Tao Yang, Di Wu, Karanjit Kalsi, and Karl H. Johansson, “Distributed Optimal Dispatch for Distributed Energy Resources over Networks with Packet Drops ”, IEEE Transactions on Smart Grid, vol. 8, pp. 3125-3137, Jun., 2017, (regular paper).
    [15] Tao Yang, Jie Lu, Di Wu, Junfeng Wu, Guodong Shi, Ziyang Meng, and Karl H. Johansson, “Dis- tributed Algorithm for Economic Dispatch over Time-Varying Directed Networks with Delays”, IEEE Transactions on Industrial Electronics, vol. 64, pp. 5095-5106, Jun., 2017, (regular paper).
    [14] Junfeng Wu, Yuzhe Li, Daniel E. Quevedo, and Ling Shi, “Improved Results on Data-Driven Power Control for State Estimation”, Systems & Control Letters, vol. 107, pp. 44-48, Sep., 2017.
    [13] Duo Han, Junfeng Wu, Huanshui Zhang, and Ling Shi, “Optimal sensor scheduling for multiple linear dynamical systems”, Automatica, vol. 75, pp. 260-270, Jan., 2017, (regular paper).
    [12] Junfeng Wu, Ziyang Meng, Tao Yang, Guodong Shi, and Karl H. Johansson, “Sampled-data Consen- sus over Random Networks”, IEEE Transactions on Signal Processing, vol. 61, no. 17, pp. 4479-4492, Sep., 2016, (regular paper).
    [11] Duo Han, Yilin Mo, Junfeng Wu, and Ling Shi, “An Opportunistic Sensor Scheduling Solution to Remote State Estimation over Multiple Channels”, IEEE Transactions on Signal Processing, vol. 64, no. 18, pp. 4905-4917, Sep., 2016, (regular paper).
    [10] Junfeng Wu, Xiaoqiang Ren, Duo Han, Dawei Shi, and Ling Shi, “Finite-horizon Gaussianity-Preserving Event-based Sensor Scheduling in Kalman Filter Applications”, Automatica, vol. 72, pp. 100-107, Oct., 2016.
    [9] Duo Han, Keyou You, Lihua Xie, Junfeng Wu, and Ling Shi, “Optimal Parameter Estimation under Controlled Communication over Sensor Networks”, IEEE Transactions on Signal Processing, vol.63, no.24, pp.6473-6485, Dec., 2015, (regular paper).
    [8] Junfeng Wu, Ling Shi, Lihua Xie, and Karl H. Johansson, “An Improved Stability Condition for Kalman Filtering with Bounded Markovian Packet Losses”, Automatica, vol. 62, pp. 32-38, Dec., 2015.
    [7] Junfeng Wu, Yuzhe Li, Daniel E. Quevedo, Vincent Lau, and Ling Shi, “Data-driven Power Control for State Estimation: A Bayesian Inference Approach”, Automatica, vol. 54, pp. 332-339, 2015.
    [6] Chao Yang, Junfeng Wu, Xiaoqiang Ren, Wen Yang, Hongbo Shi, and Ling Shi, “Deterministic Sensor Scheduling for Centralized State Estimation under Limited Communication Resources”, IEEE Trans- actions on Signal Processing , vol. 63, no.9, pp. 2336 - 2348, 2015, (regular paper).
    [5] Duo Han, Yilin Mo, Junfeng Wu, Sean Weerakkody, Bruno Sinopoli, and Ling Shi, “Stochastic Event- triggered Sensor Schedule for Remote State Estimation”, IEEE Transactions on Automatic Control, vol. 60, issue 10, pp. 2661-2675, 2015, (regular paper).
    [4] Junfeng Wu, Karl H. Johansson, and Ling Shi, “A Stochastic Online Sensor Scheduler for Remote State Estimation with Time-out Condition”, IEEE Transactions on Automatic Control, vol. 59, no. 11, pp. 3110-3116, 2014.
    [3] Junfeng Wu, Ye Yuan, Huanshui Zhang, and Ling Shi, “How Can Online Schedules Improve Com- munication and Estimation Tradeoff?”, IEEE Transactions on Signal Processing, vol. 61, no. 7, pp. 1625-1631, 2013, (regular paper).
    [2] Junfeng Wu, Qing-Shan Jia, Karl H. Johansson, and Ling Shi, “Event-based Sensor Data Scheduling: Trade-off Between Communication Rate and Estimation Quality”, IEEE Transactions on Automatic Control, vol. 58, no. 4, pp. 1041-1046, 2013.
    [1] Chao Yang, Junfeng Wu, Wei Zhang, and Ling Shi, “Schedule Communication for Decentralized State Estimation”, IEEE Transactions on Signal Processing, vol. 61, no. 10, pp. 2525-2535, 2013, (regular paper).

    SELECTED PEER-REVIEWED CONFERENCE PAPERS

    [25] Xinghan Li, Haodong Jiang, Xingyu Chen, He Kong and Junfeng Wu. "Closed-form Error Propagation on $SE_n(3)$ Group for Invariant EKF with Applications to VINS", IROS, accepted.
    [24] Yan Leng, Xiaowen Dong, Junfeng Wu, Alex Pentland, “Learning Quadratic Games on Networks”, The 37th International Conference on Machine Learning ,ICML 2020.
    [23] Jieqiang Wei, Ehsan Nekouei, Junfeng Wu, Vladimir Cvetkovic, Karl H. Johansson, “Steady-state Analysis of a Human-social Behavior Model: a Neural-cognition Perspective”, 2019 American Control Conference (ACC), July 10-12, 2019.
    [22] Kam Fai Elvis Tsang, Junfeng Wu, Ling Shi, “Zeno-Free Stochastic Distributed Event-Triggered Consensus Control for Multi-Agent Systems”, 2019 American Control Conference (ACC), July 10-12, 2019.
    [21] Yang Liu, Junfeng Wu, Ian R. Manchester, and Guodong Shi, “Gossip Algorithms that Preserve Privacy for Distributed Computation Part II: Performance Against Eavesdroppers”, The 2018 IEEE Conference on Decision and Control, Miami, USA, 2018.
    [20] Yang Liu, Junfeng Wu, Ian R. Manchester, and Guodong Shi, “Gossip Algorithms that Preserve Privacy for Distributed Computation Part I: The Algorithms and Convergence Conditions”, The 2018 IEEE Conference on Decision and Control, Miami, USA, 2018.
    [19] Takuya Iwaki, Junfeng Wu, and Karl Henrik Johansson, “Event-triggered Feedforward Control subject to Actuator Saturation for Disturbance Compensation”, The 2018 European Control Conference, Limassol, Cyprus, 2018.
    [18] Jieqiang Wei, Christos Verginis, Junfeng Wu, Dimos V Dimarogonas, Henrik Sandberg, and Karl H Johansson, “Asymptotic and finite-time almost global attitude tracking: representations free approach”, Limassol, Cyprus, 2018.
    [17] Yuchi Wu, Takuya Iwaki, Junfeng Wu, Karl Henrik Johansson, and Ling Shi, “Sensor selection and routing design for state estimation over wireless sensor networks”, The 36th Chinese Control Conference, Dalian, China, 2017.
    [16] Takuya Iwaki, Yuchi Wu, Junfeng Wu, Henrik Sandberg, and Karl Henrik Johansson, “Wireless sensor network scheduling for remote estimation under energy constraints”, The 56th IEEE Conference on Decision and Control, Melbourne, Australia, 2017.
    [15] Enoch Kung, Junfeng Wu, Dawei Shi, and Ling Shi, “On the nonexistence of event-based triggers that preserve Gaussian state in presence of package-drop”, American Control Conference, Seattle, USA, 2017.
    [14] Heng Zhang, Yifei Qi, and Junfeng Wu, “Optimal jamming power allocation against remote state estimation”, American Control Conference, Seattle, USA, 2017.
    [13] Duo Han, Junfeng Wu, Yilin Mo, and Lihua Xie, “Stochastic Sensor Scheduling for Multiple Dynamical Processes over a Shared Channel”, The 55th IEEE Conference on Decision and Control, Las Vegas, USA, 2016.
    [12] Junfeng Wu, Ziyang Meng, Tao Yang, Guodong Shi, and Karl H. Johansson, “Critical Sampling Rate for Sampled-Data Consensus over Random Networks”, The 54th IEEE Conference on Decision and Control, Osaka, Japan, 2015.
    [11] Duo Han, Keyou You, Lihua Xie, Junfeng Wu, and Ling Shi “Stochastic Packet Scheduling for Optimal Parameter Estimation”, The 54th IEEE Conference on Decision and Control, Osaka, Japan, 2015.
    [10] Junfeng Wu, Guodong Shi, Brain D. O. Anderson, and Karl H. Johansson, “Stability Conditions and Phase Transition for Kalman Filtering over Markovian Channel”, The 34th Chinese Control Conference, Hangzhou, China, 2015. (Guan Zhao-Zhi Best Paper Award).
    [9] Junfeng Wu, and Karl H. Johansson, “Peak Covariance Stability of Kalman Filtering with Markovian Packet Losses”, The 2nd IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, Hong Kong, 2014.
    [8] Junfeng Wu, Guodong Shi, and Karl H. Johansson, “Probabilistic Convergence of Kalman Filtering over Nonstationary Fading Channel”, The 53rd IEEE Conference on Decision and Control, Los Angeles, USA, 2014.
    [7] Heng Zhang, Peng Cheng, Junfeng Wu, and Jiming Chen, “Online Deception Attack Against Remote State Estimation”, The 19th IFAC World Congress, Cape Town, South Africa, 2014.
    [6] Duo Han, Yilin Mo, Junfeng Wu, Bruno Sinopoli, and Ling Shi, “Stochastic Event-triggered Sensor Scheduling for Remote State Estimation”, The 52nd IEEE Conference on Decision and Control, Florence, Italy, 2013.
    [5] Junfeng Wu, Yilin Mo, and Ling Shi, Stochastic Online Sensor Scheduler for Remote State Estimation, The 1st IEEE International Conference on Cyber-Physical Systems, Networks, and Applications, Taipei, 2013.
    [4] Junfeng Wu, Karl H. Johansson, and Ling Shi, “An Improved Hybrid Sensor Schedule for Remote State Estimation under Limited Communication Resources”, The 51st IEEE Conference on Decision and Control, Maui, USA, 2012.
    [3] Chao Yang, Junfeng Wu, Wei Zhang, and Ling Shi, “Communication Scheduling for Decentralized State Estimation”, The 12th International Conference on Control, Automation, Robotics and Vision, Guangzhou, China, 2012.
    [2] Chao Yang, Junfeng Wu, Wei Zhang, and Ling Shi, “Communication Topology Design under Limited Bandwidth”, The 20th International Symposium on Mathematical Theory of Networks and Systems, Melbourne, Australia, 2012.
    [1] Junfeng Wu, Lihua Xie, and Ling Shi, “An Innovative Packet-Splitting Approach for Kalman Filtering over Lossy Networks”, The American Control Conference, San Francisco, USA, 2011.

    Patents

  • “Localizing a Target Device Based on Measurements from a Measurement Device Array”, US
    Patent US 1135354
  • RESEARCH PROJECTS


    A Filter on the matrix Lie group: Closed-form Error Propagation on $SE_n(3)$ Group for Invariant EKF


    We derive a filter on the matrix Lie group to be applied in the VINS, which achieves accurate self-localization for autonomous applications (drones, cars, and AR/VR). This filter is an extension of Invariant EKF developed on the OPENVINS{ https://docs.openvins.com/}. In this project, we establish the closed-form formula for the error propagation for the Invariant extended Kalman filter (IEKF) in the presence of random noises and apply it to vision-aided inertial navigation. We evaluate our algorithm via numerical simulations and experiments on the OPENVINS platform. Both simulations and the experiments performed on the public EuRoC MAV datasets demonstrate that our algorithm in certain parameters settings outperforms some state-of-art filter-based methods such as the quaternion-based EKF, first estimates Jacobian EKF, etc. The techniques of choice on the parameters are worth further investigating.

    A moment in the experiment

    Boxplots of the odometric relative pose error using QEKF, FEJ, IEKF and IJ-IEKF algorithms on different segments of easy, medium, and difficult sequences in the EuRoC MAV dataset.

    The root mean squared errors evolution for the QEKF, FEJ, IEKF and IJ-IEKF algorithms on the challenging sequence MH_04_difficult.

    Low-complexity Distributed Detection with One-bit Memory Under Neyman-Pearson Criterion


    Distributed detection, which arose from multi-radar object detection, has gained its flourishing with the development of wireless sensor networks (WSNs). It is now widely applied in WSNs-based event detection, e.g., spectral sensing in cognitive radio networks, distributed intrusion detection systems, and distributed detection of human activities.
    In this project, we consider a multi-stage distributed detection scenario, where n sensors and a fusion center are deployed to accomplish a binary hypothesis test. At each time stage, local sensors generate binary messages and then upload them to the FC for global detection decision making. We suppose a one-bit memory is available at the FC to store its decision history and focus on developing iterative fusion schemes. We first visit the detection problem of performing the Neyman-Pearson test at each stage and give an optimal algorithm, called the oracle algorithm, to solve it.
    Noticing the computational inefficiency of the oracle fusion, we then propose a low-complexity alternative, for which the likelihood ratio test threshold is tuned in connection to the fusion decision history compressed in the one-bit memory. The low-complexity algorithm greatly brings down the computational complexity at each stage from O(4 n) (worst case) to O(n). We show that two algorithms exponentially converge to the same performance with the increase of time stages. In addition, the rates of convergence are proven to be asymptotically identical. The proposed low-complexity algorithm is applied in real experiments of done detection using multiple acoustic sensors. Experiment results show that the one-bit information can effectively improve detection performances.

    Proposed multi-stage distributed detection diagram

    (a) Microphone array     (b) DJI Phantom2
    Drone detection based on multiple microphones

    Localizability from Range-Difference Measurements

    In this project, we study the source localization problem based on range-difference measurements, or equivalently time difference of arrival (TDOA) measurements. TDOA is a high-precision localization technique and has various applications, ranging from the global positioning system, cellular network localization to under-water sonar positioning. In TDOA-based localization, there is a reference sensor, and for each other sensor, the TDOA measurement is obtained with respect to the reference one. By minimizing the sum of spherical least squares errors, a nonconvex constrained least squares (CLS) problem is constructed. To the best of our knowledge, there is no literature giving a complete algorithm to calculate the global minimizer of this CLS problem.
    We first derive a set of theoretical results on the existence of a global solution and its explicit characterizations. Specifically, a necessary and sufficient condition is presented for the global solutions based on the method of Lagrange multipliers, and a characterization for the uniqueness of the solutions incorporating a second-order optimality condition is derived. We then develop an efficient algorithm to solve the CLS problem based on our theoretical findings. We also propose guidelines on sensor deployment to improve localization accuracy. Simulation and real-world experiments in our system demonstrate significant localization error drop compared with existing algorithms.

         (a) Scenario 1                   (b) Scenario 2
         Simulation comparison with some algorithms

    (a) UWB-based indoor localization scenario (b) Localization result
                   Test on a public dataset

    Prescribed performance controller designed on the S2 space

    Over the last few decades, rigid body attitude control has gained many research interests afresh, booming applications such as target surveillance by unmanned vehicles, camera calibration in computer vision.
    We explore the use of attitude control algorithms for tracking fast moving UAV targets in the sky. The main challenge lies in that in such a visual servoing application, to track the UAV, learning on visual information extracted from pixels representing features in camera images, the target of interest must lie in the field of view (FOV), which makes it rather demanding in transient and steady state performance for the tracking algorithm. We develop an adaptive attitude tracking algorithm with minimal information required from the target, e.g., the orientation of a target, maintaining reduced design complexity. Analysis and experimental results show that the algorithm has superior transient and steady-state tracking performance even with consideration of quantization and saturation of servo motors.

    Coverage control on the S2 space

    This project focuses on distributed Pan Tilt(PT) camera networks, where each camera's position is fixed and attitude is variable to perform distributed sensing tasks of covering a spherical area. We model PT camera coverage utility by accounting for both limited-range sensing capability and dynamic performance. A control law is developed to regulate the cameras’ attitudes for solving the coverage optimization problem, leading to the monotonic ascent of the coverage utility encoding a collaborative sensing task. Finally, numerical simulations are performed to demonstrate the possibility of applying our algorithm in an airspace surveillance scenario on the roof of a building to detect air vehicles invasion. The simulations illustrate the effectiveness of our results. .


    Evaluation. Different moments of simulation.Left column shows four cameras with initialized states located on the building. Right columns show cameras stop at different attitudes in the end. And these time-varied red regions can lead these four cameras.

    Minimum-Degree Distributed Graph Filter Design

    Graph signal processing (GSP) offers a unified framework to process the data sets exhibiting irregular relationships between samples, which can be captured by a graph. The goal of GSP is to extend the conventional signal processing techniques (e.g., transforms, sampling, filtering) to analyze data associated with a graph representing the relational structure within the data.
    One of the cornerstones of the GSP field is graph filter (GF). In direct analog to classical time-domain filtering, GFs manipulate a signal by selectively amplifying/attenuating its graph Fourier coefficients. This renders them ideal for a wide range of tasks, ranging from signal smoothing and denoising, classification and reconstruction, consensus of multi-agent systems, segmentation, wavelet construction, and dictionary learning.
    We establish fundamental connections between local response of shifting at a node, concerned in the GSP field, and the observability of the system, investigated in control theory.
    Specifically, by introducing a notion of observable graph frequencies to a node, we show that the output signals (observations) at a node only contain the spectral components of its so-called observable graph frequencies. We use low-pass GFs as an example to illustrate the application of our proposed Algorithm 1 in computing a consensus value for a sensor network. We randomly generate an undirected graph of 7 nodes and treat the Laplacian matrix as the operator A with specific value. We compute the consensus value using the proposed node-variant GF and a minimum number of observations.

    PEOPLE

    Current


    Haoying LI(李昊颖)

    April 2022-present
    B.Eng., Zhejiang University

    Haodong JIANG (江昊东)

    September 2021-present
    B.Eng., Zhejiang University

    Bokang ZHANG (张博康)

    September 2021 - Present
    B.Eng., Zhejiang University
    Research interests: AI and Cyber privacy and security

    Junhu JIN (金俊虎)

    September 2021 - Present
    B.Eng., Zhejiang University
    Research interests: Graph signal processing

    Jianhui LI (李建辉)

    September 2020 - Present
    BEng., Zhejiang University
    Research interests: Reinforcement Learning

    Ke FANG(方科)

    Sep.2019 – Present
    MPhil., Beihang University
    B.Eng., Harbin Institute of Technology
    Research interests: Online control;Distributed control systems

    Xinghan LI(李星翰)

    September 2018 - Present
    B.Eng., Zhejiang University
    Research interests: SLAM;Filtering and Estimation on SO(3)

    Shuaiting HUANG(黄帅婷)

    September 2018 – Present
    B.Eng., Nanjing Agricultural University
    MSE, Nanjing Agricultural University
    Research interests: Distributed state estimation;Cyber security and privacy

    Guangyang ZENG(曾广扬)

    Co-supervision
    September 2017 – Present
    B.Eng., Zhejiang University
    Research interests: Distributed detection; Estimation in sensor networks
    Personal website:https://guangyangzeng.github.io/

    Past

    Kemi DING(丁克蜜)

    Visiting scholar
    Oct.2021-Dec.2021
    Assistant Professor at Southern University of Science and Technology
    Pesonal website:https://kemiding.github.io/

    Haishan ZHANG(张海珊)

    Visiting Student
    Sept.2021-Apr.2022
    Undergraduate,Nankai University

    Xingyu CHEN(陈星宇)

    Visiting Student
    Sept.2021-Jan.2022
    Undergraduate,The Australian National University

    Gallery


    WUTong Mountain

    Chuanjia Dinner Party

    Street BBQ