基于惯性/偏振光/光流的六足步行机器人自主导航方法研究

针对惯性/卫星组合导航系统易受干扰或自主性不足等问题,引入偏振光传感器和光流传感器,分别建立航向角和速度量测方程,以辅助惯性导航系统,提出了一种基于惯性/偏振光/光流的自主导航方法。同时,为实现惯性传感器、偏振光传感器和光流传感器等多传感器的融合,设计了无迹Kalman滤波器。为验证该方法的有效性,以六足步行机器人为对象开展仿真和实验验证。结果表明,在没有卫星信号源的情况下,仅依靠机器人自身感知,可实现较高精度的机器人位姿估计,实现了不依赖于卫星导航信号的自主导航,提升了导航系统的自主性。

Research on Autonomous Navigation Method of Hexapod Walking Robot Based on Inertial/Polarization Skylight/Optical Flow Integrated Navigation System

For the problems of the INS(Inertial Navigation System) /GNSS integrated navigation system being vulnerable to interference or lack of autonomy, the polarized skylight sensor and optical flow sensor are introduced to improve the integrated navigation system. The heading angle and velocity measurement equations are established, respectively, to assist the inertial navigation system. Then, an autonomous navigation method based on INS, polarized skylight sensor and optical flow sensor is proposed. At the same time, an unscented Kalman filter(UKF) is designed to fuse the multiple sensors such as inertial sensors, polarized light sensors, and optical flow sensors. In order to verify the validity of the method, the hexapod walking robot is used to carry out simulation and experimental verification. The results show that, without satellite signal sources, only relying on the robot''s own perception, higher accuracy robot pose estimation and the autonomous navigation that does not depend on satellite can be achieved, which improves the autonomy of the navigation system.