Research Experience

2024~2025 Whole-Body Model Predictive Control for Magnetic Actuated Robotic Endoscope, CUHK

  • To ensure the operation safety, we need to schedule the motion of end-effector and magnetic endoscopy simultaneously. Traditional hierarchical controller cannot achieve this goal.
  • We proposed a whole-body optimal control method, in which the endoscopy and manipulator are modeled as an integrated system.
  • To increase the computational efficiency, we proposed a partial simplification method and combined the real-time iteration framwork with the MPC.

2023~2024 Confidence-aware catching for the floating-base manipulator subject to disturbance, during the visit at Maastricht University

  • Proposed a method to realize the collision-free approach before catching and determining the boundary of the target motion based on maximum likelihood estimation.
  • Realize the multi-step prediction of the target that performs quasi-random motion in real-time based on the wavelet network. Evaluate the confidence of predictions based on the Bayesian method.
  • Catching object that performs stochastic motion by combining the confidence evaluation and whole-body nonlinear programming, taking the confidence and feasibility into catching simultaneously.

2019~23 Design and control of a manipulator-based landing assistance system for USV-UAV systems, supported by National Natural Science Foundation of China in CUHK-Shenzhen

  • Manipulator controller design for UAV tracking and catching, improving the tracking performance by designing a robust controller based on the Lyapunov method.
  • Design model predictive controller based on the joint model with delay estimation and generate reference trajectory based on the wavelet network.
  • System modeling, controller verification, and simulation based on Matlab/Simscape; field experiments of the assistance landing system.

2018~22 Design and control of a wave-driven dual-axis solar tracker, CUHK-Shenzhen

  • Feedback system design based on kinematics analysis; dynamics modeling and decoupling for lower computational complexity.
  • Wave-driven controller design based on motion prediction; sliding mode observer design for higher robustness; motion planning for lower energy cost.
  • Mechanical design and modeling of a solar tracker without actuators; reduce the required base motion for motion driving based on numerical optimization.

2015~18 Design and control of a permanent-magnet spherical motor(PMSM) for conformal printing, supported by the National High Technology Research and Development Program of China and National Natural Science Foundation of China in HUST

  • Mechanical design and modeling of a three-DoF spherical motor; optimize magnets arrangement to improve the available workload; force analysis and structure optimization based on ANSYS/Workbench.
  • Kinematics and dynamics analysis based on Euler-Lagrange equation; optimal tool path planning.
  • Non-singular terminal sliding mode controller design for robust motion control; stability analysis based on Lyapunov theory; PID controller design and parameter tuning.
  • Control cabinet design; system implementation based on Labview and CompactRIO.