题 目：A Hybrid Offline Optimization Method for Reconfiguration of Multi-UAV Formations
时 间：2022年6月28日16:00-18:00
地 点：理学院202
线 上：腾讯会议号 673-189-037
主 讲 人：李彬 教授
李彬，教授、博士生导师。2005年-2011年在哈尔滨工业大学控制科学与工程专业获得本科、硕士和博士学位，2011年在澳大利亚科廷大学获得数学与统计系获得博士学位。2012年至2014在西澳大利亚大学电气电子与计算机工程系从事研究工作（Research Associate），2014至2017年在澳大利亚科廷大学数学与统计系从事研究工作（Research Fellow）。目前，在四川大学空天科学与工程学院任教授。IEEE高级会员，中国人工智能学会智能空天专委会委员，中国航空学会导航制导与控制专委会青年委员。主要从事最优控制，数值优化，信号处理，以及无人机、卫星的自主和集群控制。共发表SCI检索期刊论文40余篇，其中包括IEEE TSP、TWC、TCOM和TAES等顶级杂志论文。主持国家自然科学基金面上和青年项目各1项目，国防基础科研课题一项，装发慧眼行动课题1项，参与JKW和重点研发等多项国家重大课题。在国际学术会议多次做大会报告和特邀报告。目前担任国际期刊Journal of Industrial Management and Optimization的编委，国内期刊《中国舰船》编委。
报告内容：Formation reconfiguration of multiple unmanned aerial vehicles (UAVs) is a challenging problem. Mathematically, this problem is an optimal control problem subject to continuous state inequality constraints and terminal state equality constraints. The first challenge is that there are an infinite number of constraints to be satisfied for the continuous state inequality constraints, which makes the problem extremely difficult to be solved. The second challenge is that the control and state are usually both been discretized. This will result in noncontinuous control input. In addition, the discretized system may not always accurately approximate the original system. In this talk, a hybrid offline optimization scheme is proposed to tackle these problems. Unlike the existing methods, the state variables are not required to be discretized and continuous control inputs can be obtained. In addition, the continuous state inequality constraints are tackled without increasing the total number of constraints. Simulation results show that the proposed hybrid optimization method outperforms the state-of-the-art method-the hybrid particle swarm optimization and genetic algorithm.
欢迎广大师生前来聆听！
