Education
Beihang University
Ph. D in Control Science and Control Engineering, advised by Prof. Lei Guo, Prof. Xiang Yu, and Prof. Kexin Guo.
Shenyuan Honors College (≤3 %).
Ph. D in Control Science and Control Engineering, advised by Prof. Lei Guo, Prof. Xiang Yu, and Prof. Kexin Guo.
Shenyuan Honors College (≤3 %).
Fall 2020 - Spring 2025
Beihang University
M. S in Control Science and Control Engineering.
M. S in Control Science and Control Engineering.
Fall 2019 - Spring 2020
Harbin Engineering University
B. S in Automation.
B. S in Automation.
Fall 2015 - Spring 2019
Appointments
Nanyang Technological University
Postdoc in the School of Mechanical and Aerospace Engineering, working with Prof. Jianfei Yang at the MARS Lab.
Postdoc in the School of Mechanical and Aerospace Engineering, working with Prof. Jianfei Yang at the MARS Lab.
Nov 2025 - now
Hangzhou Innovation Institute, Beihang University
Research Fellow.
Research Fellow.
Jun 2025 - Nov 2025
Shenzhen Taiyi Technology (SKYMAGIC)
Drone control intern.
Drone control intern.
Aug 2021 & Aug 2022
Main Awards and Honors
Outstanding graduate of Beihang University
2025
Beihang, China
Beihang, China
Academic Excellence Fund for Doctoral Candidate
Funds, ≤1 %, Project leader
Funds, ≤1 %, Project leader
2024 - 2025
Beihang, China
Beihang, China
Outstanding Research Project of Shen Yuan Honors College
Funds, Project leader
Funds, Project leader
2022 - 2024
Beihang, China
Beihang, China
2024 ICCA Best student paper award
Title： TRACE - Trajectory refinement with control error enables safe and accurate maneuvers.
Title： TRACE - Trajectory refinement with control error enables safe and accurate maneuvers.
2024
IEEE International Conference on Control and Automation
IEEE International Conference on Control and Automation
2019 CAC First Prize of Outstanding Paper
Title : Dual-disturbance observers-based control of UAV subject to internal and external disturbances.
Title : Dual-disturbance observers-based control of UAV subject to internal and external disturbances.
2019
IEEE Chinese Automation Conference
IEEE Chinese Automation Conference
Four-times Outstanding Academic Innovation Award
2022 - 2023
Beihang, China
Beihang, China
Two-times First-class Scholarship
2021 - 2022
Beihang, China
Beihang, China
China Classification Society Scholarship
2017
China Classification Society
China Classification Society
First prize in Mathematical Contest in Modeling
2016
Heilongjiang Province
Heilongjiang Province
First prize in the Chinese Mathematics Competitions
2016
China
China
National Scholarship
2016
China
China
Publications
ADAPT: Analytical disturbance-aware policy training for humanoid locomotion
We present ADAPT, an analytical disturbance-aware framework for humanoid locomotion that infers joint-level external disturbances from robot dynamics and proprioception alone, without specialized sensors, and feeds them back into policy training for robustness to pushes, payloads, and impacts.
arXiv, 2026
APEX: Adaptive policy execution for precise manipulation
APEX is a plug-and-play framework that bridges the execution gap of learned manipulation policies by reconstructing dynamically feasible references from policy outputs and adapting them online via state feedback.
arXiv, 2026
MARS policy: Multimodality only when it matters
We present MARS, a generative policy that adaptively invokes stochasticity only when behavioral branching truly matters and reverts to deterministic learning otherwise, yielding 16.67% higher success and 83.20% lower inference latency in real-world tests.
arXiv, 2026
Feedback world model enables precise guidance of diffusion policy
We close the prediction–observation loop in world models with a lightweight feedback state that corrects future predictions online, paired with action-aware guidance to emphasize controllable dimensions, reducing prediction error by up to 76.4% and improving OOD success by 30%.
arXiv, 2026
FLASH: Efficient visuomotor policy via sparse sampling
We present FLASH, an efficient visuomotor policy that replaces iterative denoising with continuous Legendre polynomial trajectories, enabling single-step inference for real-time robot control.
arXiv, 2026
Physics filtering favors the generalization of robot learning
We propose PhyFilter, which enhances both generalization and interpretability of robot learning. PhyFilter operates by correcting learning outcomes according to readily accessible physical differential structure and real-time state feedback.
Preprint, 2026
World model for robot learning: A comprehensive survey
A comprehensive survey of world models for robot learning, covering world models as policies and as simulators across recent literature.
arXiv, 2026
Action-to-action flow matching
We introduces A2A, an efficient generative paradigm that replaces noise-based initialization with action-to-action transport.
Robotics: Science and Systems (RSS), 2026
Learning-based observer for coupled disturbance
We present a learning-based observer, enabling accurate prediction of the coupled disturbance consisting of internal uncertainties and external disturbances.
IEEE International Conference on Robotics and Automation (ICRA), 2026
Optimizing control-friendly trajectories with self-supervised residual learning
The presented trajectory optimizer outputs trajectories that are friendly to the following control level.
arXiv, 2026
FORESEER: Recognize and utilize uncertainties by integrating data-based learning and symbolic feedback
We present two converged uncertainty prediction frameworks, enabling accurate prediction of two general kinds of uncertainties, respectively.
The International Journal of Robotics Research (IJRR), 2025
Feedback favors the generalization of neural ODEs
We present feedback neural networks, showing that a feedback loop can flexibly correct the learned latent dynamics of neural ordinary differential equations (neural ODEs), leading to a prominent generalization improvement.
International Conference on Learning Representations (ICLR), 2025
EVOLVER: Online learning and prediction of disturbances for robot control
We present a framework, namely EVOLVER, to mimic the bio-behavior for robotics to achieve rapid transient reaction ability and high precision steady-state performance simultaneously.
IEEE Transactions on Robotics (T-RO), 2024
Accurate high-maneuvering trajectory tracking for quadrotors: A drag utilization method
Different from standard approaches that achieve precise tracking by feedforward compensating the estimated drag, this work presents a scheme to appropriately utilize drag.
IEEE Robotics and Automation Letters (R-AL), 2022
Agile flight control under multiple disturbances for quadrotor: Algorithms and evaluation
A scheme of anti-disturbance agile flight control is developed for a maneuverable quadrotor unmanned aerial vehicle, subject to the aerodynamic drag, dynamic shift of center of gravity, and motor dynamics.
IEEE Transactions on Aerospace and Electronic Systems (T-AES), 2022
Multiple observers-based anti-disturbance control for a quadrotor UAV against payload and wind disturbance
This paper presents a multiple observers-based anti-disturbance control scheme against multiple disturbances for a quadrotor unmanned aerial vehicle.
Control Engineering Practice, 2020
Flight control for quadrotor safety in the presence of CoG shift and loss of motor efficiency
A safety control strategy based on a novel nonlinear disturbance observer and geometric control is developed for a quadrotor unmanned aerial vehicle, subject to the disturbances caused by center-of-gravity shift and loss of motor efficiency.
International Conference on Unmanned Aircraft Systems (ICUAS), 2022
Dual-disturbance observers-based control of UAV subject to internal and external disturbances
This paper presents an embedded micro loop to enhance the anti-disturbance performance for unmanned aerial vehicles.
China Automation Conference, 2019