During motor operation, the motor parameters change, which causes parameter drift. They are also affected by internal and external unknown disturbances, which lead to reduced motor control performance, poor anti-interference performance, and low robustness. A method termed ultra-local model-free predictive current control (MFPCC) has previously been proposed to solve this problem; it uses only the input and output of the system and does not involve any motor parameters, because of which it is free of problems caused by model mismatch. However, the conventional MFPCC method requires adjustment of several control parameters and the estimated value of the total disturbance of the system has a certain deviation and a large pulsation, which result in obvious chattering of the motor output, low stability, reduced anti-interference performance, and low robustness. Therefore, this paper proposes an MFPCC method based on nonlinear disturbance compensation (NDC). This method does not involve any motor parameters, and it can more accurately and stably estimate the total system disturbance, and feedforward compensation, real-time update control information, only need to adjust two control parameters, the workload is small. Simulation results show that the proposed control method has high anti-interference performance, high robustness, small output ripple, and improved dynamic characteristics and that it can estimate the system disturbance accurately and stably.
At present, the neglect of deployment friction could lead to a significant deviation in the numerical results from the real deployment dynamic characteristics of the nonconductive space tether. To solve this problem, a dedicated experimental setup for tether deployment was firstly employed to measure the deployment friction of nonconductive space tether, the dependences of the deployment friction on the tether diameter, deployment velocity and deployment angle (which is the angle between the deployment direction and the normal direction of the tether exit) were determined. The effect of deployment parameters on the dynamics in both the uncontrolled tether deployment and the station-keeping stages were evaluated based on the "dumbbell" and "extended dumbbell" models, the measured deployment frictions was applied to the models. The numerical results show that the uncontrolled nonconductive tether deployment is great limited by the deployment friction; the tether deployment capability and dynamic stability could be enhanced by increasing the initial deployment velocity (<= 2 m/s) and satellite effective mass, as well as decreasing the orbital altitude. To guarantee maximum tether deployment capability and dynamic stability, an optimal matching relationship between the tether full length and satellite total mass is given, which is of substantial importance for the uncontrolled deployment process of the nonconductive space tether system at the design stage.
A sliding mode control (SMC) strategy based on Byrnes-isidori (B-I) normal form is proposed to solve the tracking control problem of non-minimum phase (NMP) hypersonic flight vehicle (HFV) under external disturbance (ED). A nonlinear disturbance observer (NDO) is used for ED, the estimation error is guaranteed to be bounded stable. A dynamic sliding mode controller is constructed to stabilize internal dynamics, and a dynamic inverse controller is designed to stabilize the external dynamics. Aiming at stabilizing the velocity subsystem, a sliding mode controller is designed. Asymptotic stability of internal dynamics, practical finite time stability of velocity tracking error and bounded stability of external dynamic are proved by Lyapunov theory. Simulation is used to verify the effectiveness of the control strategy.
In the domain of concrete penetration, test data are often limited in quantity and unevenly distributed, which leads to poor accuracy of the machine learning-based model for predicting the depth of concrete penetration. This paper aims to improve the accuracy of the model within the constraints of limited penetration test data. In this paper, based on collecting a large amount of penetration test data, the penetration data was extended by data augmentation methods such as linear interpolation and adding Gaussian noise. The genetic algorithm and greedy algorithm were used to optimize four common machine learning models’ hyperparameters: multilayer perceptron (MLP), radial basis neural network (RBF), support vector regression (SVR), and extreme gradient boosting tree (XGBoost). The results show that using linear interpolation and adding Gaussian noise can effectively alleviate the problems of insufficient data and uneven data distribution. The average error of MLP, RBF, and XGBoost decreases by 2.7
In recent years, high-strength steel bars (HSSB), such as hot-rolled ribbed bars (HRB) and heat-treated ribbed bars (HTRB), have been promoted and used in reinforced concrete (RC) structures worldwide. There is still some controversy in existing research regarding the crack development process, failure modes, and the applicability of current code-specified calculation formulas for high-strength RC beams. The study of the mechanical properties of high-strength RC beams is of great significance in promoting the development of construction engineering science. In order to investigate the flexural behavior of high-strength RC beams, four-point bending static loading tests were carried out on 11 groups of RC beams with different parameters (reinforcement strength, concrete strength, and reinforcement ratio). The test results indicate that the crack development process and failure mode of high-strength RC beams exhibit obvious bending failure characteristics. Increasing reinforcement strength and reinforcement ratio significantly improves the bearing capacity of the beams, while the impact of increasing concrete strength on the beam's capacity is relatively small. The applicability of the calculation formulas in the current codes was analyzed, it was found that the yield moment formula in the current code accurately predicts the yield moment of high-strength RC beams, and the formula for calculating the maximum crack width was modified by proposing the maximum crack width adjustment factor k omega. Subsequently, the failure process of highstrength RC beams was simulated using explicit dynamic finite element software LS-DYNA. Parameterized analyses were conducted to broaden the research scope. The ultimate moment prediction factor ku was proposed and a formula that can accurately predict the ultimate moment of high-strength RC beams was established.
In a complex electromagnetic environment, the satellite signals and communication links of unmanned aerial vehicle (UAV) systems are often unstable or susceptible to interference. In this paper, a cooperative navigation method is proposed by combining GNSS/INS integrated navigation technology with visual navigation technology, based on the effective information collected by onboard sensors. This method fuses the relative navigation information of multiple UAV (multi-UAV) digital image maps with the absolute navigation information obtained by each UAV. Navigation message transmission rules are designed to reduce the amount of information exchanged between UAVs. By deploying the Kalman filter algorithm across the UAVs, the computational requirements are reduced and computational efficiency is improved. Finally, relevant experiments are conducted in a simulated mission scenario, and the results demonstrate that the proposed method satisfies the design requirements in terms of the accuracy of position and velocity information in cooperative navigation.
针对机动目标定位的精度提升问题,给出两种典型任务场景下机动目标定位的在线协同航迹规划方法.建立测向交叉目标定位模型,基于几何精度因子确定影响定位精度的因素,分析影响要素对定位精度的影响规律;对于低速机动目标定位,以几何精度因子为评估指标,提出一种基于变曲率杜宾斯曲线的在线协同航迹生成方法,实现协同定位最优构型的快速航迹生成;对于高速机动目标协同定位问题,建立面向目标定位精度的最优控制模型,提出一种基于内点法的罚函数协同航迹规划方法;进行数字与半实物仿真验证.研究结果表明:在两种典型任务场景下,目标协同定位精度提升36.9％和23.5％,验证了新方法的有效性,新方法对于集群协同目标定位具有工程应用价值.
针对现有路径规划方法对地形特征考虑不足的问题,以无人履带车辆为研究对象,提出一种基于可通行度估计的路径规划方法.基于卷积长短期记忆(Conv LSTM)网络,从连续轨迹上提取激光雷达点云的空间特征和时间关联特征,融合车辆运动特征,估计地形可通行度.基于地形可通行度,改进A算法的节点扩展方式和代价函数,输出满足无碰撞约束和低可通行代价的离散路点;使用无梯度迭代平滑算法减小路径松弛度和可通行度代价;再使用三次B样条曲线对离散路径进行拟合,输出平滑参考路径.以参考路径建立Frenet坐标系,构建基于可通行度代价的安全走廊,在满足无碰撞约束、低可通行度代价的前提下,在走廊内生成满足车辆运动学约束的平滑路径.试验结果表明,所提出的方法能够充分考虑地形特征,提升路径规划结果的稳定性和可通行性.
为研究具有复杂地形的未知战场环境中的全光化无人机集群作战概念,提出基于智能体建模与仿真的全光化无人机集群效能和体系贡献率评估方法.根据全光化无人机集群作战特点建立评估指标体系,给出各作战单元的模块化Agent模型.对超低空巡逻打击任务开展仿真,分析全光化无人机感知性能、通信性能、集群规模和集群配置对集群生存率、任务完成率、战损比和体系贡献率的影响.研究结果表明:全光化无人机能为作战体系带来新能力,有效提升体系作战效能;集群规模越大、全光化无人机占比越高,则体系作战效能越强,但全光化无人机性能提升带来的贡献并不明显.
双侧电机耦合驱动履带车辆单侧电机发生故障如果不及时采取措施,极易导致车辆偏驶,甚至出现安全问题.为了保证单侧电机故障模式下的车辆安全,开展单侧电机故障模式下车辆制动避障安全控制研究.基于实车采取的一侧发生故障、另一侧及时处于故障模式的控制方式进行车辆安全性分析,提出一种双侧电机耦合驱动履带车辆单侧电机故障模式下车辆安全控制策略并通过RT-LAB半实物实时仿真验证.研究结果表明:该控制策略能够按照驾驶员意图,在单侧电机故障模式下实现不同车速下车辆不同相对转向半径的转向控制,而且面对连续的避障需求,可以稳定转向,保证履带车辆的安全.
作战效能预测对武器装备体系从建设、生产到实战的全过程都具有重要意义.在Stac-king集成学习模型的基础上,优化模型对数据的交叉验证方式,针对原有模型次级学习器输入向量较为稀疏的问题,为次级学习层的输入增加多项式特征和经主成分分析法降维后的各项作战仿真数据指标(原始数据),形成一种改进Stacking集成学习模型的装备体系作战效能预测方法.以合成营攻占某一阵地的作战效能预测为例,验证该方法的有效性.
针对多导弹在三维空间以期望角度协同攻击机动目标的问题,提出一种三维领弹-从弹时间协同制导律.根据弹目相对运动关系建立领弹-从弹三维非线性协同制导模型,无需小角度假设;在视线法向和侧向上,基于2阶滑模控制理论和设计的有限时间收敛滑模面,分别设计领弹-从弹三维角度控制制导律,在提高系统收敛速度的同时抑制了抖振现象;在从弹视线方向上提出时间协同制导律,创新性地将领弹-从弹协同制导问题转化为2阶多智能体一致性跟踪控制问题,充分利用导弹间的信息交互,在实现从弹和领弹时间协同的同时避免传统的剩余时间估计造成的误差问题;同时对新算法分别进行严格的Lyapunov稳定性证明.仿真结果表明:新的三维领弹-从弹时间协同制导律可以有效控制领弹和从弹在三维空间以期望角度高精度协同攻击机动目标.
针对装甲车辆综合传动装置外接油管动态特性恶化导致的失效问题,研究不同车速动态激励下的外接油管动态特性和结构强度变化规律,提出外接油管有效优化改进措施.对综合传动装置上典型外接润滑油管和供油油管系统进行动态特性分析.利用试验测试获取油管关键部位的载荷激励,进行瞬态动力学仿真和理论动力学建模,探究关键参数对油管系统动态特性的影响规律.提出动态特性的优化改进方案,油管系统关键部位的振动幅值消减15％以上.
近年来,任务卸载作为保障无人集群高效协同作战的关键技术之一,正成为研究热点.任务卸载旨在克服单平台算力不足、能量有限等约束,将计算任务卸载到边缘网络的服务器上进行处理,以达到降本增效的目的.以无人集群辅助的天地一体化协同侦察为作战场景,考虑战时复杂多变的电磁环境以及集群组网拓扑时变性,利用Lyapunov优化把长期任务卸载解耦为在线马尔可夫决策过程.为解决混合动作空间收敛难、学习效率底的问题,结合凸优化和多智能体深度确定性策略,分层求解功率分配和任务分配问题,提出数据-模型双层优化驱动的多智能体强化学习卸载决策算法.数值实验结果表明,新算法能够根据时变的战场环境自适应调整智能体任务卸载策略,达到提升传统算法性能和优化复杂多维目标的目的.
针对分布式电驱动车辆多动力源耦合作用和高度非线性造成的动力学控制难题,以7自由度整车动力学模型为预测模型,以粒子群优化-蚁群融合算法为优化方法,提出一种基于粒子群优化-蚁群融合算法的模型预测转矩协调控制策略,并搭建了仿真实验和实车试验平台,进行了多种工况试验.试验结果表明,新提出的转矩协调控制策略能够根据试验工况调整控制模式,实现动力性、经济性和操纵稳定性的综合最优控制效果.
包络线控制起源于航空航天工业,它提供了飞行状态的安全保障和机动边界,为飞行器控制带来了良好效果.基于8×8多轴分布式驱动无人车辆和包络线方法核心思想,提出一种将车辆推向极限的整车动力学控制器.通过建立轮胎滑移圆提出一种新的方法以用于评估车辆驱动力状态,并将轮胎滑移状态与车辆"g-g"图相结合,用来实现无人驾驶状态下逼近车辆操纵能力极限,发挥车辆动力性能与灵活性能,同时确保在轨迹跟踪时的跟踪精度,精准高效地完成平台任务.考虑外界环境不确定扰动与因素变化对极限状态下车辆稳定性影响,基于车辆横向动力学模型的稳定特性分析,获得不同条件下稳定域相平面,并探索其变化机理、归纳数学描述表达式.通过对车辆稳定相平面的分析,提出以车辆横摆力矩为输出的稳定保持控制器.针对上层控制器驱动力与横摆力矩的输出,设计下层转矩分配控制策略,通过冗余执行器的最优分配实现整车性能发挥.整车集成控制策略部署于一辆8×8原型试验车辆,在越野路面上进行多项科目测试,试验结果表明:在高速条件下,无人车在轨迹跟踪中具有更好的动力性能和安全性能.
未来战场的多样化对特种无人车辆的环境适应性提出了更高的要求.为满足特种无人车辆的多工况使用、高机动能力与低成本研制等要求,采用多工况关联设计与轻量化优化的思路对无人车桁架车身结构进行优化设计.考虑到多工况的车身结构设计变量多、设计空间大,面临仿真次数过多的问题,提出基于多工况关联的车身结构轻量化优化方法,利用设计变量区间缩减策略减小设计空间;引入高斯过程代理模型替换仿真分析实现结构设计方案性能的快速评估;结合遗传算法实现方案的优化.实验结果表明,最终优化方案在通过车身刚度强度和模态等多学科性能仿真验证的情况下,质量比初始方案降低14.12％,比只用高斯过程优化设计的方案降低8.87％.
随着军用地面无人系统研究的深入,单一的地面无人机动平台或任务载荷很难满足现代战场的需求,只有任务载荷和机动平台协同发展,地面无人系统才能在战场中真正形成战斗力.为进一步推动任务载荷与机动底盘协同技术的发展,综述了搭载任务载荷军用地面无人系统的发展背景、研究现状及技术特点,分别从多层次多维度的环境建模、基于多模态数据的通行度估计、基于多智能体协同建模的协同规划控制优化方法三方面对其关键技术进行阐述,总结了相关的研究框架和重点,并对搭载任务载荷军用地面无人系统未来的发展方向进行了展望.
协同定位和环境感知技术是无人集群实现自主导航的基石,但受制于大规模无人集群系统小型个体平台的计算、载荷、带宽等资源所限,诸多相关技术难以实际部署应用.为实现资源约束下大规模无人集群的精准定位与环境感知,提出一种基于半直接法的轻量化协同视觉SLAM算法,设计融合光流法和直接法的半直接特征点跟踪方法,采用集中式双向通讯策略,使得大规模无人集群系统在面对通讯干扰和延迟时拥有较高的容错率,同时兼具准确性和快速性.基于Eu-RoC数据集和实际物理环境对算法开展对比实验,结果表明:新算法的实时性能平均提升60％,显著优于其他基于特征法的协同视觉SLAM算法;在丢包率小于40％以及通讯延迟低于0.1 s的低质量通讯环境中,新算法定位精度更高、鲁棒性更强.
针对武器装备技术的发展对防空作战带来的挑战,提出一种基于高功率微波(HPM)武器系统与中近程防空武器协同作战目标分配设计模型.该目标分配模型首先设定作战场景,考虑HPM武器系统的软、硬杀伤效能,与中程防空导弹、近程防空导弹和末端近防炮协同作战,分析HPM武器系统及中近程防空武器的拦截效率,使用模拟退火算法给出目标分配模型求解方法,得出拦截效率最大化的防空武器分配方案.仿真实验结果表明,相比于蒙特卡洛算法和粒子群优化算法,模拟退火算法能够在更短的时间内寻找到拦截效率更高的解,有效解决协同作战中的目标分配问题,为HPM武器系统融入多武器协同防空作战系统奠定基础.
