摘要
为适应自主驾驶车辆的高精度、高频率与高可靠性的导航要求,提出了一种机器视觉/数字地图/CP-DGPS共同辅助SINS的智能车辆组合导航方法,建立了组合导航系统的滤波模型。该滤波模型的量测信息不仅包括GPS与SINS形成的位置与姿态观测信息,还包括机器视觉/数字地图/SINS形成的横向偏差观测信息。通过对SINS的多重冗余辅助,使得导航系统具备容错能力。仿真结果表明,该组合导航系统能为智能车辆提供其空间位置、速度、加速度与姿态角等众多导航信息,并具有100Hz的高频输出、厘米级的导航精度和容错性能,当GPS较长时间中断时,通过SINS/视觉/数字地图的组合仍能为智能车辆提供可靠的导航数据。
To satisfy the navigation demands such as good accuracy, high frequency and high reliability for autonomous intelligent vehicle, a new multi-sensor integrated navigation technique was studied. According to the characteristics of Intelligent Vehicle＇s navigation sensors, an integrated navigation method was proposed in which SINS is aided by Vision/Digital Map/CP-DGPS. The navigation filter model was then set up. In this filter, the observation information includes two parts： one is the position and attitude information formed by SINS/GPS while the other is the lateral deviation error information formed by Vision/Digital Map/SINS. With SINS aided redundantly, the navigation system has the fault-tolerant ability. Simulation results show that the proposed navigation system can provide abundant navigation information（including the position, velocity, acceleration and attitude etc.） at 100 Hz with cm-level navigation accuracy and good fault-tolerant performance. Even when GPS is continuously interrupted for a period, The Intelligent Vehicle can still obtain reliable navigation information by SINS/Vision/Digital Map integration.
出处
《中国惯性技术学报》
EI
CSCD
2007年第3期316-321,共6页
Journal of Chinese Inertial Technology
基金
江苏省汽车工程重点实验室开放基金资助项目(QC200603)
江苏省交通科学研究计划项目(06C04)
关键词
智能车辆
自主驾驶
组合导航
信息融合
intelligent vehicle
autonomous driving
integrated navigation
information fusion
