Ke-Bo Li

Dr. Ke-Bo Li

orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
9 February 2024 in Drones
Verified
Journal Article
Open access
PDF file
JTJin Tang
Jin Tang
YLYangang Liang
Yangang Liang
KLKe-Bo Li
Ke-Bo Li
Traditional unmanned aerial vehicle path planning methods focus on addressing planning issues in static scenes, struggle to balance optimality and real-time performance, and are prone to local optima. In this paper, we propose an improved deep reinforcement learning approach for UAV path planning in dynamic scenarios. Firstly, we establish a task scenario including an obstacle assessment model and model the UAV’s path planning problem using the Markov Decision Process. We translate the MDP model into the framework of reinforcement learning and design the state space, action space, and reward function while incorporating heuristic rules into the action exploration policy. Secondly, we utilize the Q function approximation of an enhanced D3QN with a prioritized experience replay mechanism and design the algorithm’s network structure based on the TensorFlow framework. Through extensive training, we obtain reinforcement learning path planning policies for both static and dynamic scenes and innovatively employ a visualized action field to analyze their planning effectiveness. Simulations demonstrate that the proposed algorithm can accomplish UAV dynamic scene path planning tasks and outperforms classical methods such as A*, RRT, and DQN in terms of planning effectiveness.

23 March 2021 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
Hyo-Sang ShinHyo-Sang Shin
Hyo-Sang Shin
KLKe-Bo Li
Ke-Bo Li

1 April 2024 in Journal of Guidance, Control, and Dynamics
Verified
Journal Article
Closed access
HLHaojian Li
Haojian Li
YLYuanhe Liu
Yuanhe Liu
KLKe-Bo Li
Ke-Bo Li
YLYan’gAng Liang
Yan’gAng Liang
Aiming at the impact-angle-control guidance with field-of-view constraint in the three-dimensional space, a novel analytical prescribed performance guidance law is proposed. First, a new prescribed performance function is designed in the relative range domain, and the common drawbacks of most existing prescribed performance functions, such as large overshoot, time dependency, and infinite-time convergence, are avoided. Second, the three-dimensional biased pure proportional navigation guidance frame is selected. The biased term is derived based on the prescribed performance control theory and the optimal error dynamic method, for realizing the terminal impact-angle constraint. After that, the analytical solution of missile lead angle is derived under some linearization approximations, and two methods are respectively developed for obtaining the range of the guidance gain and updating its value to guarantee the satisfaction of the field-of-view constraint during the guidance process. Moreover, for the proposed guidance law and methods, the tracking error could be ensured to remain in the predefined performance boundaries. Finally, numerical simulations are conducted to verify the effectiveness of the new theoretical findings.

24 October 2019 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Open access
KLKe-Bo Li
Ke-Bo Li
Hyo-Sang ShinHyo-Sang Shin
Hyo-Sang Shin
Antonios TsourdosAntonios Tsourdos
Antonios Tsourdos

30 August 2019 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Open access
KLKe-Bo Li
Ke-Bo Li
Hyo-Sang ShinHyo-Sang Shin
Hyo-Sang Shin
Antonios TsourdosAntonios Tsourdos
Antonios Tsourdos
MTMin-Jea Tahk
Min-Jea Tahk
The capturability of two-dimensional (2D) pure proportional navigation (PPN) guidance law against lower-speed arbitrarily maneuvering target for homing phase had been thoroughly analyzed by using the nonlinear output regulation (NOR) method before. However, due to the complexity of the three-dimensional (3D) relative kinematics, the NOR method has not been applied to the capturability analysis of 3D PPN, which leads to the capturability discrepancy of 2D PPN and its 3D extension. Thanks to the 3D relative kinematic equation between the missile and target established in the rotating line of sight (LOS) coordinate system, the capturability of 3D PPN against the lower-speed arbitrarily maneuvering target for the homing phase is restudied by extending the NOR method of 2D PPN to the 3D space. The necessary and sufficient condition for the missile guided by 3D PPN to intercept this type of target is obtained. It is proven that the capturability of 3D PPN is identical with that of 2D PPN.

1 December 2023 in Journal of Guidance, Control, and Dynamics
Verified
Journal Article
Closed access
HLHaojian Li
Haojian Li
YLYuanhe Liu
Yuanhe Liu
KLKe-Bo Li
Ke-Bo Li
YLYan’gAng Liang
Yan’gAng Liang
To address the issue of intercepting maneuvering targets at a specific time, a polynomial guidance method for impact-time control is proposed in this paper. Based on the relative virtual framework and the classical differential geometry curve theory, such method is divided into two parts: 1) the design of the relative trajectory-length-control (RTLC) guidance law against virtual stationary targets, and 2) the design of the prediction algorithms based on the guidance law or its characteristics. The former realizes RTLC, and the latter establishes the relationship between the desired relative trajectory length-to-go and the desired time-to-go, thus implementing impact-time control. Furthermore, based on the analytical properties of the guidance law in the relative arc-length domain, its performance, characteristics, and allowable impact time are analyzed. Finally, the effectiveness of the proposed guidance method and the validity of the theoretical findings are verified by numerical simulations results.