项目名称: 基于不确定性集合在线建模的旋翼飞行机器人高机动飞行控制方法研究
项目编号: No.61203334
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 自动化学科
项目作者: 宋大雷
作者单位: 中国科学院沈阳自动化研究所
项目金额: 24万元
中文摘要: 高机动飞行控制方法研究对旋翼飞行机器人广泛应用具有重要意义。飞行机器人动力学复杂,表现出欠驱动和强耦合。非结构的随机扰动又使其高机动动力学呈现出不确定性,无法基于参考模型设计飞行控制器。现有可应对动力学不确定性的基于静态摄动界分析的鲁棒控制方法过于保守,不适用于响应速度要求较高的高机动飞行控制。采用估计方法动态地对不确定性在线建模,并基于此在线模型进行控制器设计,是解决鲁棒控制保守性的一条有效途径。然而,现有估计方法多在噪声分布已知条件下采用参数值的估计来建立在线模型,在实际未知有色噪声条件下,无法跟踪高机动飞行中突变的不确定性。鉴于此,本课题拟提出一种基于不确定性集合在线建模的鲁棒控制方法,不同于常规不确定参数值的估计,此方法拟通过在线估计包含动力学不确定性的最小集合,并基于此估计集合提出在线鲁棒控制算法,兼具稳定性与快速性,适用于旋翼飞行机器人高机动飞行控制,具有理论和应用研究价值。
中文关键词: 集员估计;旋翼飞行机器人;不确定性建模;飞行控制;
英文摘要: Rotor Unmanned Aerial Vehicle (RUAVs) applications raise the performance requirements of aggressive flight in terms of controllability. As a result of RUAV's complex dynamics, RUAVs are under actuated and unstable; meanwhile, nonstructural environment and internal random disturbances generate system uncertainty in aggressive flight; therefore, it is unable to design a flight controller based on an accurate dynamic model. However, it is quite conservative in robust control of RUAV's uncertainty effectively by using static Perturbation Bounds analysis, which reduces the control responsibility in aggressive flight. One effective approach to solve the conservativeness of robust control is to dynamically online modeling the uncertainty using estimate method and design the controller based on this online model. Yet, current estimation methods commonly use numerical estimation to build online model with the noise distribution known. It is unable to track the uncertainty in dynamics during aggressive flight with unknown distribution of the noises. Given this, this project proposes an online modeling robust control method based on uncertainty set. Unlike common uncertain parameter estimation, this method estimates the minimal dynamic uncertainty with online modeling, and then establishes online robust control algorithm b
英文关键词: set-membership estimation;fly robot with rotory wings;uncertainty modeling;flight control;