With the rapid development of computer vision and machine vision, deep learning based methods have achieved good results in field target detection, recognition, tracking. However, for ship detection recognition on sea surface, rate is greatly affected by uneven distribution horizontal vertical features ships different sizes ships. In order to improve accuracy real-time performance, this paper proposed a algorithm YOLO V5, which feature extraction process was merged with GhostbottleNet...

This paper develops an adaptive course controller for time-varying parametric uncertain nonlinear ships with completely unknown bounded control coefficient. The proposed design method does not require any a priori knowledge of the sign designed autopilot can guarantee regulation ship to prescribed accuracy and global uniform ultimate boundedness all signals in closed-loop system. effectiveness presented has been demonstrated simulation involving 45 m length.

The response model of podded propulsion unmanned surface vehicle (USV) is established and identified; then considering the USV has characteristic high speed, course controller with fast convergence speed proposed. idea MMG separate modeling used to establish three‐DOF planar motion USV, simplified as a model. Then based on field experiments, parameters are obtained by method system identification. Unlike ordinary ships, advantages small size, so needs strong robustness. Based theory...

This paper presents an adaptive backstepping sliding mode tracking control method for underactuated unmanned surface vehicle. The controller is designed by combining and technology. virtual law used to replace the strategy of position error, transformed into velocity control, which can effectively avoid input saturation singular value problem existing in design traditional law. neural shunt model solve differential explosion caused technology based on theory introduced compensate uncertainty...

Predicting the trajectory of flying objects with spin is a challenge but an essential task in many fields, especially military and sports. Robots playing ping-pong very good platform to validate prediction method. Various vision systems have been proposed, only position information was used most cases, which limits their capability predict spinning ball. Based on fact that ball's motion can be separated into translation movement respect center, this paper proposes novel system provide both...

In order to simultaneously address the issues of ship operating area limitation, unknown time-varying disturbances, immeasurable speed, dynamics, and input saturation, this paper investigates position-constrained dynamic positioning output feedback control, taking thruster system dynamics into account. Firstly, a barrier Lyapunov function (BLF) is utilized limit position inside system’s acceptable working range error. Second, set total disturbance, which made up disturbances further handled...

This paper investigates the path following control problem for an underactuated unmanned surface vehicle (USV) in presence of dynamical uncertainties and time-varying external disturbances. Based on fuzzy optimization algorithm, improved adaptive line-of-sight (ALOS) guidance law is proposed, which suitable straight-line curve paths. On basis information provided by LOS, a three-degree-of-freedom (DOF) dynamic model USV has been used to design practical controller. The controller designed...

The course control system of an unmanned surface vehicle (USV) is a complex with time-varying, nonlinear and multi-disturbances. Traditional PID controllers often cause overshoot concussion in USV control. Intelligent Track Guiding Control (ITGC) introduced to suppress the controller design. On this basis, fuzzy applied optimize parameters construct adaptive guide (FC-ITGC). controlled object identified recursive least squares method. Considering presence or absence interference, FC-ITGC...

For ping-pong playing robots, observing a ball and predicting ball's trajectory accurately in real-time is essential. However, most existing vision systems can only provide position observation, do not take into consideration the spin of ball, which very important competitions. This paper proposes way to observe estimate real-time, achieve an accurate prediction. Based on fact that spinning motion be separated global movement respect its center, we construct integrated system two motions...

This paper addresses three interrelated problems concerning the path following of a podded propulsion unmanned surface vessel (USV), namely, modeling USV, model parameter identification and guidance law design. Based on hypotheses simplification, degree freedom (DOF) manipulating mathematical group (MMG) models USV are reduced to response model, then, series field experiments (turning experiment zig‐zag experiment) is carried out collect corresponding data. On this basis, parameters obtained...

Deep learning has achieved a great success in both visual and acoustic recognition classification tasks. The accuracy of many state-of-the-art methods have surpassed that human beings. However, the field robotics, it remains to be big challenge for real robot master high-level skill using deep methods, even though can easily learn task from demonstration, imitation practice. Compared Go Atari games, this kind tasks is usually continuous state space action space, which makes value based...

The interface formation and properties of composite materials are very important for the preparation materials, bonding state charge transfer between atoms in have a particularly significant effect on formation. In this work, first-principles calculation method was used to study adsorption behavior molecular dynamics copper (111) surface H-terminated diamond, energy adhesion work Cu were calculated. results show that is not sensitive diamond surface, small at four high symmetry positions,...

Due to unmanned surface vehicle (USV) has the features of underactuated and nonlinear, it control problems have always been difficulty in field study course is basis its various controls. So this paper, nonlinear USV model identified model's validity verified by rotation test Z firstly. Then, for uncertainty parameters random disturbance, backstepping method employed design controller. Moreover, order eliminate static error, integral term introduced algorithm. Theoretical results on...

In order to solve the problem of precise trajectory tracking control for a quadrotor in presence external disturbance and system model parameter uncertainty, nonlinear controller based on sliding mode is designed. this paper, mathematical QBall2 considering characteristics actuator established virtual mapped motor, then PID method Finally, simulation result shows correctness effectiveness designed controller.

This paper focuses on the accurate path following problem for an underactuated unmanned surface vehicle with uncertainties and environment disturbances. The proposed scheme can be divided into guidance loop control loop: in loop, law is utilized to calculate desired yaw angle, as well estimating unknown currents sideslip angle simultaneously; a novel robust developed by enhanced trajectory linearization (TLC) technology, nonlinear tracking differentiator (NTD), sigmoid function based...

The research on the collision between a spinning ball and table, as general issue for trajectory analysis of spinning-flying objects, is important but complicated. Traditionally, it studied either by simplifying process black box in which ball's motion state changes abruptly before after collision, or assuming that coefficient restitution friction are simply piecewise linear even constant during whole process. In this paper, we first analyze mechanism dynamics conclude from mathematical...