| 引用本文: | 郑峰婴,王峰,甄子洋,许梦园,范涛.先进布局无人机多目标自适应概率引导控制分配[J].控制理论与应用,2022,39(12):2366~2376.[点击复制] |
| ZHENG Feng-ying,WANG Feng,ZHEN Zi-yang,XU Meng-yuan,FAN Tao.Control allocation of multi-objective adaptive probabilistic guidance for advanced layout unmanned aerial vehicle[J].Control Theory and Technology,2022,39(12):2366~2376.[点击复制] |
|
| 先进布局无人机多目标自适应概率引导控制分配 |
| Control allocation of multi-objective adaptive probabilistic guidance for advanced layout unmanned aerial vehicle |
| 摘要点击 819 全文点击 261 投稿时间：2021-07-14 修订日期：2022-07-15 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2021.10625 |
| 2022,39(12):2366-2376 |
| 中文关键词 先进布局 自适应概率引导 多目标粒子群算法 控制分配 飞行控制 |
| 英文关键词 advanced layout adaptive probability guidance multi-objective particle swarm optimization algorithm (MOPSO) control allocation flight control |
| 基金项目 国家自然科学基金项目(61803200, 61973158), 装备预研重点实验室基金项目(6142220180304)资助. |
|
| 中文摘要 |
| 针对先进布局无人机多操纵面冗余的控制分配问题, 提出一种基于自适应概率引导的混合多目标控制分配方法. 首先, 根据冗余舵面操纵特性, 建立带约束的舵面动态效能模型, 提出精度需求不同的混合多目标优化指标. 随后, 为了综合平衡各目标寻优精度与求解速度提出基于自适应概率引导的多目标粒子群控制分配方法. 该方法根据各目标最优值与期望精度差值构建自适应概率函数, 依概率选择全局最优解, 引导种群向各目标期望精度方向精细搜索以提升算法解算精度, 减少无用搜索以提高求解速度; 同时, 根据收敛性指标增加变异因子, 避免算法陷入局部最优. 最后, 仿真验证该方法可有效处理舵面耦合及非线性特性, 减少能耗损失, 实现操纵面多目标控制分配, 使得无人机快速平稳跟踪控制指令. |
| 英文摘要 |
| A hybrid multi-objective control allocation method based on the adaptive probability guidance is proposed to solve the problem of redundant control allocation of advanced layout unmanned aerial vehicle (UAV). Firstly, according to the control characteristics of redundant control surfaces, a constrained dynamic effectiveness model of control surfaces is established, and the hybrid multi-objective optimization indexes with different precision requirements are proposed. Then, in order to comprehensively balance the optimization precision and solution speed of each target, a multi-objective particle swarm control allocation method based on the adaptive probability guidance is proposed. The method constructs an adaptive probability function according to the difference between the optimal value and the expected precision of each target, selects the global optimal solution according to probability, guides the population to search finely in the direction of the expected precision of each target, improves the solution precision of the algorithm, and reduces useless search to improve the solution speed. At the same time, according to the convergence index, the variation factor is added to avoid the algorithm falling into local optimum. Finally, the simulation results show that the method can effectively deal with the coupling and nonlinear characteristics of the control surface, reduce the energy loss, and realize the multi-objective control allocation of the control surface, so that the UAV can track the control command quickly and smoothly. |
|
|
|
|
|