Junkai Tan portrait

Junkai Tan谭浚楷

HKPFS Ph.D. Fellow PolyU AAE

Scholar CV PDF

WeChat · Kai

WeChat QR code for Junkai Tan

Scan to connect on WeChat.

Learning, Inference & Control

I am an HKPFS Ph.D. Fellow in the Department of Aeronautical and Aviation Engineering at The Hong Kong Polytechnic University (PolyU), advised by Prof. Wen-Hua Chen and affiliated with RCLAE.

Current Focus

I am exploring how Prof. Chen’s dual control for exploration and exploitation (DCEE) connects with active inference, studying how robots gather useful information while pursuing a task. My longer-term interests are world models and the links between autonomous control and cognitive neuroscience.

Recent Highlights

News archive

  1. Policy–safety separation under polygonal constraints: accepted by Science China Technological Sciences.

  2. Stackelberg attack–defense control: accepted by IEEE/CAA JAS as a regular paper.

  3. Resilient critic-learning control: available in IEEE TII Early Access.

  4. XJTU Graduate with Honor (2025–2026).

  5. Best Student Paper Award, IEEE ICAIS & ISAS · XJTU Qihang Role Model Nominee.

Selected Research

All publications

Earlier focus: safe reinforcement learning, prescribed-time control, and differential games, with applications to UAV–UGV coordination and human–UAV interaction.

  1. Paths around polygonal obstacles, with the complete legend for individual runs, mean path, worst-case run, start, goal, and safety boundaries.

    SCTS2026 · Accepted

    Separating control policy learning from safety enforcement under polygonal constraints

    Separates nominal policy learning from geometry-based safety enforcement for systems operating among polygonal obstacles.

    Junkai Tan, S. Xue, H. Cao, D. Li, B. Chen

    Science China Technological Sciences, 2026.

  2. Dual-clock control architecture showing the learning-dominant and control-dominant phases, actor, critic, and flexible input saturation.

    IEEE TSMC2026

    Dual-Clock Guaranteed-Time Actor-Critic Optimal Control with Flexible Input Saturation

    Assigns separate deadlines to actor–critic learning and state response while accounting for actuator limits.

    Junkai Tan, S. Xue, H. Cao, B. Chen

    IEEE Transactions on Systems, Man, and Cybernetics: Systems, 56(10): 5952–5965, 2026. IEEE Xplore

  3. Three-dimensional convergence paths in the prescribed-time numerical example with T_p = 2.0 seconds, including the time plane, initial-state paths, and complete legend.

    IEEE TNNLS2026

    Flexible Prescribed-Time Optimal Control with Adaptive State-Input Constraint Bounds via Actor–Critic Learning

    Combines actor–critic optimal control with adaptive state–input constraint bounds and a prescribed response time.

    Junkai Tan, S. Xue, H. Cao, B. Chen

    IEEE Transactions on Neural Networks and Learning Systems, 37(11): 6412–6425, 2026. IEEE Xplore

  4. Complete resilient critic-learning architecture connecting cyber, communication, and physical layers, with the controller, disturbance, FDI attack, and DoS attack paths.

    IEEE TII2026

    Fixed-Time Convergent Resilient Critic-Learning Control With Asymmetric Input-State Constraints Under Hybrid FDI–DoS Attacks

    Studies resilient critic-learning control under hybrid FDI–DoS attacks and asymmetric input–state constraints.

    Junkai Tan, S. Xue, Q. Guan, X. Zheng, H. Cao, B. Chen

    IEEE Transactions on Industrial Informatics, Early Access, pp. 1–12, 2026. IEEE Xplore

  5. Indoor UAV tracking experiment with reference and proposed trajectories, start and goal markers, and the full plot legend.

    IEEE TIE2026

    Fixed-time stochastic learning from human-UAV interaction with state-input constraints

    Learns from human–UAV interaction while handling state and input constraints.

    Junkai Tan, S. Xue, Q. Guan, Z. Guo, H. Cao, B. Chen

    IEEE Transactions on Industrial Electronics, 73(2): 3071–3082, 2026. IEEE Xplore

  6. Leader and follower docking trajectories from multiple initial positions, with start and end markers and both position axes.

    IEEE/CAA JAS2026

    Fixed-time hierarchical game-based unmanned aerial-ground vehicle docking control

    Coordinates UAV–UGV docking through a hierarchical game-based control scheme.

    Junkai Tan, S. Xue, Z. Guo, H. Cao, B. Chen

    IEEE/CAA Journal of Automatica Sinica, 13(5): 1242–1244, 2026. IEEE Xplore

  7. Complete planning-and-tracking scene showing the quadrotor, ground vehicle, and marked virtual obstacle in the indoor experiment.

    IEEE TASE2025ESI Highly Cited

    Hierarchical safe reinforcement learning control for leader-follower systems with prescribed performance

    Pairs obstacle-aware leader planning with prescribed-performance follower tracking in an interconnected UAV–vehicle system.

    Junkai Tan, S. Xue, H. Li, Z. Guo, H. Cao, B. Chen

    IEEE Transactions on Automation Science and Engineering, 22: 19568–19581, 2025. IEEE Xplore

Research software

ADP-MATLAB

From data to feedback.

8 runnable examples MIT License

Explore the project GitHub
50 recorded rollouts signal-flow illustration
Fifty recorded closed-loop phase trajectories of the ADP-MATLAB normalized pendulum, with state coordinates x one and x two.
Inside the experiment

The normalized pendulum uses a data-identified Koopman model and a value critic with 3 quadratic + 197 anchored tanh features. Policy iteration fits the critic’s output weights; the resulting feedback is frozen for these 50 rollouts.

V̂(x) = βᵀφ(x) u(x) = −ĝ(x)ᵀ∇V̂(x) / (2R)

The pulses illustrate signal flow, not measured neuron activations. The plot replays stored states from the 7 September 2026 run, sampled every 0.05 s here (every fifth original sample); no new trajectories or control values are generated.

Critic implementation ↗Experiment record ↗Displayed data & provenance ↗

Education

2026 – present
Ph.D. student, Aeronautical and Aviation EngineeringThe Hong Kong Polytechnic UniversityAdvisor: Prof. Wen-Hua ChenHKPFS & PolyU Presidential Fellowship
2023 – 2026
M.E., Electrical EngineeringXi’an Jiaotong UniversityAdvisors: Prof. Hui Cao & Dr. Shuangsi XueStudent Pacesetter · National & Guodian Nanzi Special Scholarships
2019 – 2023
B.E., Electrical EngineeringXi’an Jiaotong University

Academic Service

Full record
300+ peer reviewsRecorded on ORCID

Including IEEE TAC, IEEE TCNS, IEEE TASE, IEEE TSMC: Systems, and IEEE L-CSS.

Conference reviewer
IEEE CDC · ACC
Session chair
IEEE ICARM 2023

Contact

Open to collaboration on safe learning, autonomous control, and low-altitude robotics.

junkai.tan@connect.polyu.hk Department of Aeronautical and Aviation Engineering
The Hong Kong Polytechnic University
Hung Hom, Kowloon, Hong Kong Campus directions

Research figure

Open original image