Kai QU

Dr. Kai QU

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 December 2025 in Renewable Energy
Verified
Journal Article
Closed access
KQKai QU
Kai QU
Shuangsi XueShuangsi Xue
Shuangsi Xue
XZXiaodong Zheng
Xiaodong Zheng
DYDapeng Yan
Dapeng Yan
HCHui Cao
Hui Cao

20 August 2025 in Neurocomputing
Verified
Journal Article
Closed access
SXShang Xiang
Shang Xiang
KQKai QU
Kai QU
YTYing Tian
Ying Tian
Shuangsi XueShuangsi Xue
Shuangsi Xue
HCHui Cao
Hui Cao

23 October 2025 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
Shuangsi XueShuangsi Xue
Shuangsi Xue
JTJunkai Tan
Junkai Tan
TNTiansen Niu
Tiansen Niu
KQKai QU
Kai QU
HCHui Cao
Hui Cao
BCBadong Chen
Badong Chen
This study proposes an integrated control framework that combines prescribed performance backstepping with reinforcement learning (RL) optimization for quadrotor unmanned aerial vehicles (QUAVs) operating under disturbances. The key innovation lies in enforcing guaranteed transient and steady-state tracking performance while achieving optimal control behavior through online learning. A systematic two-stage control architecture is developed: An outer-loop position controller incorporating prescribed performance bounds handles the underactuated dynamics, complemented by an innerloop attitude controller ensuring desired orientation tracking. The inner-loop controller employs critic–actor neural networks (NNs) to approximate optimal solutions of the Hamilton-Jacobi-Bellman (HJB) equation, while the outer-loop controller optimizes position tracking performance through Hamilton-Jacobi-Isaacs (HJI) equation solutions. Rigorous stability analysis establishes uniformly ultimately bounded (UUB) convergence for all tracking errors and NN weight estimation errors using Lyapunov theory under the prescribed performance constraints. The efficacy of the proposed framework is validated through comprehensive numerical simulations, demonstrating exceptional tracking precision with position errors under prescribed performance.

25 November 2025 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
Verified
Journal Article
Closed access
Shuangsi XueShuangsi Xue
Shuangsi Xue
ZGZihang Guo
Zihang Guo
JTJunkai Tan
Junkai Tan
KQKai QU
Kai QU
HCHui Cao
Hui Cao
BCBadong Chen
Badong Chen
This work investigates the formation tracking issue for multirobot manipulator end-effectors under input constraints. A distributed formation control law is designed to guarantee the finite-time boundedness of tracking errors within the framework. To estimate the significant bias of dynamics discovered during practical multirobot collaborative manipulation tasks, a bias radial basis function neural network (RBFNN) is integrated, along with a designed adaptive updating law for expeditious approximation. In addition, an anti-windup compensator within a finite-time framework is specifically introduced to mitigate the input saturation issue arising from torque limitations in joint actuators. Finally, the system’s semi-global practical finite-time boundedness (SGPFTB) is rigorously established through Lyapunov theory. Five planar manipulators are employed in comparative computational experiments to validate the feasibility of the presented control strategy.

3 February 2026 in Neurocomputing
Verified
Journal Article
Closed access
KQKai QU
Kai QU
Shuangsi XueShuangsi Xue
Shuangsi Xue
XZXiaodong Zheng
Xiaodong Zheng
HCHui Cao
Hui Cao

5 January 2026 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
Shuangsi XueShuangsi Xue
Shuangsi Xue
ZDZhenxiang Ding
Zhenxiang Ding
JTJunkai Tan
Junkai Tan
KQKai QU
Kai QU
HCHui Cao
Hui Cao
DLDongyu Li
Dongyu Li
This paper presents a fixed-time synchronized reinforcement learning-based control (pFxT-SRLC) framework for spacecraft attitude systems that simultaneously addresses four critical challenges: predetermined convergence time guarantees, synchronized multi-axis coordination, energy optimization, and practical constraint handling. Our integrated framework combines a fixed-time super-twisting observer with adaptive switching strategy, a practical fixed-time synchronized controller with singularity avoidance, and an actor-critic reinforcement learning structure for near-optimal energy performance. The proposed method ensures all state variables converge to desired neighborhoods simultaneously within predetermined time bounds independent of initial conditions, while the constrained learning algorithm maintains stability throughout the optimization process. A second-order command filter handles input saturation constraints and reduces control chattering for practical implementation. Theoretical analysis proves fixed-time synchronized stability with explicit convergence time bounds, and establishes energy optimality through Hamilton-Jacobi-Bellman approximation. Simulation results demonstrate that the proposed pFxT-SRLC achieves comparable tracking accuracy to conventional fixed-time methods while reducing energy consumption and generating significantly smoother control signals.