无人车/行人动态杆臂估计与协同导航方法
Dynamic Lever Arm Estimation and Collaborative Navigation Method for Unmanned Vehicles/Pedestrians
杨子傲 1曲麒富 2左健文 1唐嘉乔 1张佳荷1
作者信息
- 1. 北京理工大学自动化学院,北京 100081;导航、制导与控制技术教育部工程研究中心,北京 100081
- 2. 中国航天系统科学与工程研究院,北京 100037
- 折叠
摘要
针对复杂动态环境下微惯导自身存在较大的器件噪声导致的定位误差随时间快速累积的问题,充分发挥了无人车/行人相对运动学的模型作用,研究了无人车/行人惯性基协同导航方法,实现了复杂运动条件下惯导系统误差抑制与精度提升.首先,建立了无人车/行人"速度+姿态"匹配模型;然后,充分考虑了行人的运动特点,建立了无人车/行人相对运动模型,并进行了两者间动态杆臂估计;最后,提出了无人车/行人协同导航方法,并开展了高可信的软件在环仿真实验.实验结果表明,在无人车/行人相对运动约束下,行人微惯导导航的定位精度大幅提升,支持不依赖卫星的无人车/行人相对导航与协同导航.
Abstract
In response to the problem of rapid accumulation of positioning errors over time caused by significant de-vice noise of micro inertial navigation systems under complex dynamic environments,the model of relative kinematics be-tween unmanned vehicles and pedestrians is fully utilized.A collaborative navigation method based on unmanned vehicle/pedestrian inertial bases is investigated to achieve the error suppression and accuracy improvement of inertial navigation sys-tems under the conditions of complex motion.Firstly,a"velocity+attitude"matching model for unmanned vehicle/pedes-trian is established.Then,taking into full consideration the motion characteristics of pedestrians,a relative motion model of unmanned vehicle/pedestrian is established,and the dynamic lever arm estimation between the two is performed.Finally,a collaborative navigation method for unmanned vehicle/pedestrian is proposed and highly reliable software in the loop simu-lation experiments are conducted.The experimental results show that he positioning accuracy of pedestrian micro inertial navigation is greatly improved under the constraint of relative motion between unmanned vehicle/pedestrian,and supports satellite independent relative navigation and collaborative navigation of unmanned vehicle/pedestrian.
关键词
协同导航/相对运动/误差补偿/传递对准Key words
collaborative navigation/relative motion/error compensation/transfer alignment引用本文复制引用
出版年
2024
NETL
NSTL
万方数据