Search

Aiming at the probe docking control problem in aerial recovery, this article proposes a neural-adaptive probe dynamics direct control algorithm with preassigned docking performance for the unmanned aerial vehicle (UAV) to be recovered, even under multiple ambient airflow disturbances and initial docking deviations. The affine nonlinear dynamics of the probe in the inertial axis are formulated, together with the UAV's 6-DOF dynamics, for the convenience of the direct nonlinear docking control design. To compensate UAV's unmeasurable nonlinear dynamics with better approximation property and lower computational burden, the minimal learning parameter (MLP) technique based echo state network (ESN) approximators are constructed by employing the system state approximation errors to adaptively update neural weights learning. Then, to guarantee the preassigned docking performance required by the recovery safety, a finite-time prescribed performance control algorithm incorporated backstepping control law is developed to achieve the desired docking. Both the docking trajectory and forward docking velocity are not only always constrained within the preassigned performance constraints but also finely controlled to achieve accurate rendezvous with the target drogue. The closed-loop stability is discussed with Lyapunov analysis. Numerical simulations are conducted to validate the performance of the docking control algorithm. [ABSTRACT FROM AUTHOR]

This article contrives a neural-adaptive constrained controller of the cable towed air–ground recovery system subject to terrain obstacles, unmeasurable cable tensions, trailing vortex, wind gust, and actuator saturation. In air–ground recovery system modeling, the towed vehicle's nominal 6 DOF affine nonlinear dynamics and the cable system's finite links-joints dynamics are formulated. To achieve accurate air–ground recovery under terrain obstacles, an asymmetric barrier Lyapunov function-based flight controller of the towed vehicle is proposed, by transforming the terrain obstacles into time-varying constraints on the vehicle's trajectory. Then, to approximate the towed vehicle's lumped unknown dynamics caused by the unmeasurable cable tensions and airflows, several echo state network (ESN) approximators are established for velocity and attitude subsystems. By using the state approximation errors-based neural weights learning strategy and minimal learning parameter technique, these ESNs possess better transient behaviors and lower online computational burden. Furthermore, the actuator saturation is automatically monitored and released, by incorporating a specially designed auxiliary compensating system into the angular rate control law for compensation. The stability of the closed-loop system is analyzed. Finally, numerical simulations under two air–ground recovery scenarios are performed to demonstrate the performance of the proposed controller. [ABSTRACT FROM AUTHOR]

In this article, considering the performance limitations of the sensors used in real missions, the cooperative dynamic fuzzy perimeter surveillance (CDFPS) problem and its fluid-based solution framework are proposed for the first time. First, the two-dimensional concentration diffusion distribution function is used to imitate an irregular dynamic perimeter. Then, according to the onboard sensor resolution ratio, a fuzzification process is applied for the perimeter area. On this basis, the detected fuzzy perimeter area is modeled with virtual obstacles (VOs). Considering an obstacle set composed of VOs, fixed obstacles and agents, we transform the CDFPS problem into a unified obstacle avoidance problem for the first time and a fluid-based path planning method is designed based on restrained interfered fluid dynamical system and Lyapunov guidance vector field. To quantify the surveillance effect, two indexes for the CDFPS problem are proposed: the perimeter neutrality index and the phase angle uniformity index. Around these proposed indexes, the attractive/repulsive effects and the multiagent cooperative mechanism are constructed. Finally, the receding horizon control is introduced to further improve the surveillance effect. Simulations show that the proposed framework can drive agents to persistently track the perimeter and establish relatively stable phase angle offsets between adjacent agents. [ABSTRACT FROM AUTHOR]

To enhance the mission success rate, the multi-agent system (MAS) often adopts a formation form to shuttle over complex terrain. Compared with obstacle avoidance in other multi-agent coordinated missions, multi-agent formation (MAF) obstacle avoidance is more challenging because agents must both avoid obstacles and subsequently return to their intended positions in the original formation at a uniform speed. Aiming at the three-dimensional complex obstacle environment, the fluid methodology is extended to the application of the formation issue for the first time. In this paper, an MAF obstacle avoidance framework built upon model-based interfered fluid dynamical system (MIFDS), which is the most representative algorithm in fluid methodology, is proposed. The proposed MIFDS is modified to be more in accordance with the kinematic performance of agents than traditional IFDS by introducing the kinematic model and constraints. Moreover, on the basis of MIFDS and the virtual leader formation structure, a novel original fluid speed adjustment strategy in MIFDS, which is the core to solve the formation issue, is designed to balance obstacle avoidance and formation reconfiguration. Finally, receding horizon control is introduced to further guarantee the performance of obstacle avoidance and motion safety. Simulations are used to demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In this study, two types of three-dimensional (3-D) flight-based energy management strategies for solar-powered unmanned aerial vehicle (SUAV) long-endurance target tracking are proposed. In SUAV dynamic modeling, we consider the influence of wind on SUAV kinematics. Aiming at the complexity of the long-endurance tracking, namely the long flight time and space spans, the influences of the wind, variations of temperature, air density and solar cell photovoltaic efficiency, and a finite battery capacity are considered in our modified SUAV energy model. Then, considering the mission flexibility, we design two types of energy management strategies that comprehensively utilize the solar energy, gravitational potential energy, and wind to address different situations. The proposed strategies can be divided into several stages in chronological order according to the solar irradiation conditions, electrical energy storage conditions, and relative motion relationships among the SUAV, target, and wind. We then design respective strategies for these stages. The core of our proposed energy management strategies is to design the corresponding objective functions in each stage and the switching conditions between adjacent stages. Finally, we verify our proposed strategies by comparison with the traditional 2-D tracking strategy in a 24-h target tracking mission, and the results show the validity of our proposed strategies. [ABSTRACT FROM AUTHOR]