Monocular vision-based finite time continuous control method for water surface aircraft

Abstract

The invention discloses a monocular vision-based finite time continuous control method for a water surface vehicle. The method belongs to the field of visual control of the water surface vehicle, andcomprises the following steps: acquiring an actual image and an expected image of a visual feature point through a monocular camera, recovering a pose error of the water surface vehicle by utilizing ahomography decomposition technology, and constructing a kinematics and dynamics model of a servo system; designing an NITSMC surface by utilizing the speed adjustment error so as to adapt to the unmatched visual uncertainty; designing an FOD, and observing matching disturbance in dynamics; introducing an exponential power reaching rate, and designing a continuous finite time monocular vision servo controller, so that continuous control input is ensured, and finite time convergence of the whole servo system can be realized. The method solves the control problem of a monocular vision servo system with non-matching uncertainty, improves the convergence rate and the control precision of adjustment errors, can effectively avoid feature point loss and vision servo failure caused by disturbance,and is particularly suitable for actual engineering requirements.

Description

technical field [0001] The invention relates to the field of visual control of water surface vehicles, in particular to a limited-time continuous control method for water surface vehicles based on monocular vision. Background technique [0002] As a portable, cheap and efficient sensor, the monocular camera has gradually become a powerful tool for marine environment perception and self-state estimation, which greatly enhances the autonomy and intelligence of surface vehicles. [0003] Young-Ho Kim et al. defined the control error in the image and designed a PID controller to realize the docking between aircraft; Wang Jian et al. designed a position-based visual servo controller using the Backstepping idea to realize aircraft formation control ; Combining trajectory tracking and visual measurement, Lefeber et al. designed a progressively stable trajectory tracking controller; without considering the nonlinear and time-varying disturbance of the system, Wang Ning et al. combin...

AI Technical Summary

Problems solved by technology

[0011] S2: Use the Euler-Lagrangian rotation matrix to transform the speed of the aircraft into an inertial coordinate system, and design a non-singular integral terminal sliding surface based on this variable to solve the adaptive problem of non-matching visual uncertainties;

Images