一种无人机自主避障与目标追踪方法

针对无人机自主避障与目标追踪问题，以深度Q网络(DQN)算法为基础，提出一种多经验池深度Q网络(MP DQN)算法，使无人机避障与追踪的成功率和算法的收敛性得到优化。更进一步，赋予无人机环境感知能力，并在奖励函数中设计了方向奖惩函数，提升了无人机对环境的泛化能力以及算法的整体性能。仿真结果表明，相较于DQN和双重DQN(DDQN)算法，MP DQN算法具有更快的收敛速度、更短的追踪路径和更强的环境适应性。     

横列式双倾转旋翼机控制系统研究CSCD

针对横列式双倾转旋翼机平飞加速过渡过程中的标称速度轨迹跟踪问题,提出了通过升力占比划分过渡阶段,实现标称轨迹的工程化设计。基于经典PID理论,分别设计了各过渡阶段各通道控制律和控制参数,提出了总距余弦分配的操纵分配方法,并解决了过渡过程中严重的纵垂向耦合问题。仿真结果表明,该纵向控制策略在强干扰下仍可以实现对象飞行器的平稳过渡。     

Periodic event-triggered formation control for multi-UAV systems with collision avoidance

In this paper, periodic event-triggered formation control problems with collision avoidance are studied for leader–follower multiple Unmanned Aerial Vehicles (UAVs). Firstly, based on the Artificial Potential Field (APF) method, a novel sliding manifold is proposed for controller design, which can solve the problem of collision avoidance. Then, the event-triggered strategy is applied to the distributed formation control of multi-UAV systems, where the evaluation of the event condition is continuous. In addition, the exclusion of Zeno behavior can be guaranteed by the inter-event time between two successive trigger events have a positive lower bound. Next, a periodic event-triggered mechanism is developed for formation control based on the continuous event-triggered mechanism. The periodic trigger mechanism does not need additional hardware circuits and sophisticated sensors, which can reduce the control cost. The stability of the control system is proved by the Lyapunov function method. Finally, some numerical simulations are presented to illustrate the effectiveness of the proposed control protocol.     

Improving aircraft aerodynamic performance with bionic wing obtained by ice shape modulation

For Unmanned Aerial Vehicles (UAVs) with limited electrical power to achieve effectively anti-/de-icing at the leading edge of the wing, a strategy of ice shape modulation was proposed. Isolated simulated ice shape pieces printed by 3D printing technology are mounted on a NACA0012 finite wing model, and its lift/drag coefficients and suction-side velocity fields are measured by the six-component force balance and the Particle Imaging Velocimetry (PIV), respectively. The ratio of the spanwise length of a single ice shape piece to chord length and the spanwise length of the non-icing area between the two adjacent single ice shape pieces are defined as dimensionless ice shape length (w/c) and dimensionless modulation ratio (w/λ), respectively. The results indicate that for a fixed w/λ, the wing lift coefficient first increases and then drops with increasing w/c, and a peak value exists when w/c is between 0.1 and 0.2. The lower the w/λ is, the higher the wing lift coefficient will be. The periodical variation of the flow separation area along the spanwise direction is attributed on the one hand to the acceleration effect of the flow field in the non-icing area which reduces the separation area, and on the other hand to the cross-flow caused by the streamwise vortices from the non-icing area to the icing area which promotes the mixing of the flow field (similar to vortex generators). The obtained modulation law is verified through flight tests and provides guidance for the use of ice shape modulation scheme for UAVs that cannot be completely anti-/de-icing under severe weather conditions.     

基于DJI精灵4多光谱版无人机的红树林精细分类

本文基于DJI（大疆）精灵4多光谱版无人机能够同时获取地表可见光照片和多光谱影像的特点，提出了使用摄影测量技术构建冠层高度模型（CHM）的方法，并结合多光谱影像应用于红树林分类。将上述方法应用于广东省湛江红树林国家级自然保护区高桥镇研究区，结果表明：基于无人机摄影测量技术构建的红树林CHM与多光谱影像结合后能够大幅提高红树林的分类精度，可替代传统的机载激光雷达数据；DJI精灵4多光谱版无人机数据获取效率高，能够实现高精度的红树林分类，在红树林监测业务工作中具有推广应用价值。     

考虑UAV性能约束的变速度自主避障算法研究

动态不确定环境下,为有效地提升UAV执行任务的安全性和可靠性,在UAV自主避障过程中考虑了自身的性能约束条件.在速度障碍圆弧法的基础上,考虑UAV的法向和纵向加速度约束范围,研究了速度障碍圆弧随速度矢量变化的规律,提出了一种考虑UAV性能约束的变速度自主避障算法.该算法在考虑自身性能约束条件下,为实现UAV对威胁障碍的避碰提供了更大的避碰裕度.最后,对考虑UAV性能约束的变速度自主避障算法进行了验证,仿真结果证明了算法的有效性和可行性.     

基于Q-Learning的深度神经网络自适应退避策略

针对无人机自组织网络,结合Q-Learning和深度神经网络,提出一种自适应退避策略,以提高基于竞争的MAC协议通信性能.以Matlab为仿真平台,仿真比较了自适应退避策略与二进制指数退避策略的性能.     

美国地下领域无人系统发展现状及启示

随着科学技术的高速发展,地下领域(如天然洞穴网络、人造隧道和城市地下环境等)逐渐成为人类研究的热点,利用无人系统在地下等复杂场景执行探索、搜索与搜救等任务仍面临着许多挑战.深入分析了美国地下领域无人系统技术发展的模式框架,梳理了快速轻量自主、机械天线和地下挑战赛等DARPA典型研发项目的 研究目标与现状,提炼出地下领域无人系统需要解决的关键技术,在此基础上归纳总结了美国地下领域无人系统的发展趋势.最后,结合国内发展情况提出了地下领域无人系统的发展启示.     

水面无人艇集群编队控制技术综述

随着海上作业任务的日渐复杂以及水面无人艇控制理论的发展，单水面无人艇航行控制能力的局限性逐渐显现，集群编队控制技术得到了广泛关注。针对水面无人艇集群编队控制技术，分析了当前国内外应用情况与技术研究现状，重点从典型编队队形适用性、编队控制系统结构形式、编队协同控制方法、编队控制通信技术及异构条件下协同控制技术等方面分析了水面无人艇集群编队控制技术的研究进展。最后，进行了总结和展望，以期为水面无人艇集群编队控制技术的研究提供有益的参考。     

Beam tracking method based on reconfigurable intelligent surface for obstructed communication

This paper proposes a new reconfigurable intelligent surface based three-dimensional beam tracking method to solve the beam tracking problems for the unmanned aerial vehicle with obstacles in communication channels. The proposed beam tracking method can not only regulate the reconfigurable intelligent surfaces to achieve the beam tracking of the obstructed communications, but also optimize the transmission efficiency of the communication. Firstly, a reconfigurable intelligent surface is proposed, which can correlate the transmission signals by adjusting the phase-shift matrix. Meanwhile, a new communication channel is constructed according to the reconfigurable intelligent surface, which consists of two parts. The first one is the channel between the unmanned aerial vehicle and the reconfigurable intelligent surface, the other is the channel between the reconfigurable intelligent surface and the ground base station. Note that the transmission performance of the communications can be optimized by adjusting the phase shift of each uniform linear array on the reconfigurable intelligent surface. Then, a new beam tracking method for the unmanned aerial vehicle with obstructed communications is proposed on the basis of the reconfigurable intelligent surface. By proposing the mixed genetic algorithm, the estimation accuracy of the azimuth and elevation angles is improved to enhance the performance of the beam tracking. Finally, the simulations are provided to verify the effectiveness of the proposed three-dimensional beam tracking method with reconfigurable intelligent surface.