A fast finite-time convergent guidance law with nonlinear disturbance observer for unmanned aerial vehicles collision avoidance

2019 ◽  
Vol 86 ◽  
pp. 204-214 ◽  
Author(s):  
Ning Zhang ◽  
Wendong Gai ◽  
Maiying Zhong ◽  
Jing Zhang
Author(s):  
Chaoyuan Man ◽  
Rongjie Liu ◽  
Shihua Li

In this paper, a nonlinear suboptimal guidance system is presented for the missile targeting an unknown arbitrary target. An integrated quadratic performance index is minimized in this guidance law, and the whole design is based on the exact 3D nonlinear missile-target dynamics without any linearization. Considering that the Hamilton–Jacobi–Bellman equation of a nonlinear system is quite difficult to be solved, the [Formula: see text] method is used to obtain the approximate solution without complicated online computations. Moreover, the target accelerations are regarded as the unknown disturbances, and the robustness against the target maneuvering and the external disturbances is enhanced by introducing the feedforward compensation based on the nonlinear disturbance observer. In addition, no priori knowledge like the time-to-go is needed in this suboptimal guidance law. Simulation studies show that the proposed composite guidance system can guarantee that the missile intercepts the arbitrary maneuvering target with satisfied performance.


Author(s):  
Bin Ren ◽  
Yao Wang ◽  
Jiayu Chen ◽  
Silu Chen

Abstract Robotic manipulators are complex and dynamic nonlinear mechanical systems subject to numerous uncertainties, such as payload variations, frictions, and unmodeled dynamics. To mitigate the uncertainty caused by these disturbances and minimize the tracking errors of the controllers, this study proposed a finite time tracking-based controller (FTC) that embeds a nonlinear disturbance observer (NDO) and a second-order sliding mode modifier (SOSM). The NDO was incorporated to compensate for the system's global bounded uncertainty and the SOSM employed a robust nonsingular terminal sliding-mode modifier to stabilize the controller. The theoretical analysis showed that the tracking error could quickly converge in finite time. Simulation on a typical robotics manipulator demonstrated the practical appeal of the proposed scheme.


Sign in / Sign up

Export Citation Format

Share Document