Visual inertia odometer pose estimation method based on sparse direct method

Abstract

The invention discloses a visual inertia odometer pose estimation method based on a sparse direct method. The method comprises the following steps of: firstly, giving a large rotation and translationto a carrier, and initializing a VO and an IMU installed on the carrier; constructing an N-layer image pyramid, and adjusting gray value of the whole image pyramid; maintaining a sliding window with aplurality of key frames by adopting a nonlinear least square method based on the sliding window; for each image frame, judging whether the image frame is a key frame or not according to the pose estimation of the image frame, if yes, adding the current frame into the sliding window for nonlinear optimization, marginalizing the image frame with the minimum contribution to the window in the window,and calculating and updating a priori information matrix and a priori residual error; and if the current frame is not a key frame, directly moving the current frame out of the sliding window, and maintaining the state variable in the original sliding window. According to the invention, the VIO system has better robustness to illumination.

Description

technical field [0001] The invention belongs to the technical field of Simultaneous Positioning and Mapping (SLAM). Background technique [0002] The inertial measurement unit (IMU) measures the three-axis angular velocity and acceleration of the carrier at a high frequency and calculates the attitude. It can provide short-term high-precision motion estimation, but it will be affected by factors such as zero bias, temperature, and vibration. error problem. Visual odometry (VO) can obtain more accurate camera rotation and camera translation through the read image information, and it will not drift when stationary, but it is susceptible to image occlusion, lighting environment, etc., resulting in tracking loss. Since visual sensors and inertial sensors are very complementary, visual and inertial sensors are fused to obtain a visual-inertial odometry (VIO). VIO is a special SLAM system, which has a wide range of applications in many fields, such as robotics, unmanned driving,...

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: In order to solve the problems of low calculation accuracy in the prior art, the present invention provides a visual-inertial odometer pose estimation method based on the sparse direct method

Images