Navigation Method Based on Iterative Extended Kalman Filter Fusion of Inertial and Monocular Vision

Abstract

The invention discloses a navigation method based on iterative extended Kalman filter fusion inertia and monocular vision. The method is specifically as follows: a monocular camera and an inertial measurement unit are installed on a carrier, and a message filter in ROS is used to realize the monocular camera and monocular vision. The time stamp of the inertial measurement unit is synchronized, and the pose change between the two frames of images before and after is calculated, and the position, speed, rotation and other change information obtained by the inertial measurement unit within the corresponding time are calculated, and the position obtained by the inertial measurement unit is calculated. , speed and rotation are used as the state variables of the system, and the pose change information obtained by the visual sensor is used as the observation to establish the system equation. And through an iterative extended Kalman filter method, the information obtained by the two sensors is fused to realize the real-time state estimation and navigation of the carrier. The present invention can maintain higher precision in the process of long-term real-time positioning and navigation, and has the advantage of constant computational complexity between frames.

Description

technical field [0001] The invention relates to the technical field of navigation, in particular to a navigation method based on iterative extended Kalman filter fusion inertial and monocular vision. Background technique [0002] In recent years, navigation-related instruments and equipment have made breakthroughs, and the performance and accuracy of the equipment have been greatly improved. However, the navigation method implemented by a single sensor still has certain performance limitations. In order to meet the high-performance navigation requirements, integrated navigation methods have received extensive attention and development in recent years. The integrated navigation method combines a variety of navigation sensors, and uses the information measured by multiple sensors to compensate each other for their respective limitations to achieve high-precision navigation and enhance the robustness of the system. [0003] Inertial navigation is a comprehensive technology, wh...

AI Technical Summary

Problems solved by technology

The tightly coupled method greatly increases the amount of calculation due to the addition of image feature information to the state vector, and its computational complexity has a linear relationship with the range of image features, and its scalability is poor.

Although the accuracy of loose coupling is slightly worse than that of tight coupling, its computational complexity is much smaller than that of tight coupling.

Images