基于视觉导航的动态航迹优化  

Dynamic route optimization based on visual navigation

在线阅读下载全文

作  者:吴成良[1] 曹云峰[2] 庄丽葵[2] 谢也 丁萌[3] WU Cheng-liang CAO Yun-feng ZHUANG Ei-kui XIE Ye DING Meng(College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210000, China College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 210000, China College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 210000, China)

机构地区:[1]南京航空航天大学自动化学院,江苏南京210000 [2]南京航空航天大学航天学院,江苏南京210000 [3]南京航空航天大学民航学院,江苏南京210000

出  处:《电子设计工程》2017年第3期14-18,共5页Electronic Design Engineering

摘  要:针对基于视觉导航的航迹规划技术存在的两个问题:第一,作为一类存在有限测量量程的导航系统,视觉导航系统需要目标始终出现在摄像头的视野内;第二,当目标始终在摄像头的光轴延长线附近时,视觉算法能够更有效率并且更准确的解决位姿估计问题;文中首先将航迹优化问题转化为最优控制问题,结合飞行器运动动力学特性,设计最优控制问题的约束与最优指标。然后,文中利用伪谱法将连续时间最优控制问题转化为离散时间非线性规划问题,并在此基础上研究了基于滚动时域的动态规划方法,最终减少视觉导航误差对航迹优化的不良影响。As to the two key problems of path planning technology based on vision navigation: first, as a kind of limited measuring range navigation system, the visual navigation system requires that target should be in the camera view; Second, it is more efficient and accurate for visual algorithm to solve The pose estimation problem when the target is always near the optical axis of camera extension cord. Route optimization problem is firstly transformed into the optimal control problem in this paper and the optimal control problem of constraint and optimal index is designed by combining with the aircraft motion dynamics characteristics. Then, continuous time optimal control problem is transformed into discrete time nonlinear programming problem by utilizing the method of pseudo-spectral. While the dynamic planning based on rolling time domain method is researched to reduce the bad influence of visual navigation error on route optimization.

关 键 词:视觉导航 最优控制 动态规划 GAUSS伪谱法 滚动时域 

分 类 号:TP39[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象