Kanghui He
Beihang University
9 Papers
5 Citations
Kanghui He is an academic researcher from Beihang University. The author has contributed to research in topics: Computer science & Engineering. The author has an hindex of 1, co-authored 5 publications.
Chat about Author
Papers
Learning-Based Trajectory Tracking and Balance Control for Bicycle Robots With a Pendulum: A Gaussian Process Approach
TL;DR: In this paper , a learning-based control framework for a class of underactuated bicycle robots with an active pendulum attached to the rear body as a balancer is presented, which deals with the uncertainties of changing road environment, unmodeled dynamics, and external disturbances by virtue of Gaussian process and disturbance cancellation on control device.
28
Anti-disturbance dynamic inversion backstepping control for uncertain pure-feedback systems via multiple extended state observers
TL;DR: In this paper, a novel backstepping approach was proposed by combining extended state observers with dynamic inversion controllers, with high gain properties both on the observers and controllers, the resulting closed-loop system presented relatively fast convergence.
8
Composite deep learning control for autonomous bicycles by using deep deterministic policy gradient
Kanghui He,Chaoyang Dong,An Yan,Qingyuan Zheng,Bin Liang,Qing Wang +5 more
- 18 Oct 2020
TL;DR: The results indicate that the bicycle controlled by the proposed composite deep learning based control strategy can follow along a predetermined trajectory while maintaining balance.
6
Cascade Integral Predictors and Feedback Control for Nonlinear Systems with Unknown Time-varying Input-delays
TL;DR: A cascade integral high-gain predictor is proposed to estimate the future state with a distinctive structure that can handle unknown delays and eliminate the “peaking phenomenon” during the transient period.
3
Approximate Dynamic Programming for Constrained Piecewise Affine Systems with Stability and Safety Guarantees
TL;DR: In this paper , an alternative approach based on approximate dynamic programming (ADP), an important class of methods in reinforcement learning, was proposed for optimal control of constrained piecewise affine (PWA) systems.
1