Visual-inertial tightly coupled integrated navigation method based on firefly swarm optimization pf

Abstract

The invention relates to a vision-inertia tightly-coupled integrated navigation method, in particular to a vision-inertia tightly-coupled integrated navigation method based on a glowworm swarm optimization PF (Particle Filter). The invention improves the real-time performance and navigation accuracy of the vision-inertia tightly-coupled integrated navigation method. The vision-inertia tightly-coupled integrated navigation method based on the glowworm swarm optimization PF is implemented by the following steps: S1, mounting a strapdown inertial navigation system and a binocular visual odometeron a carrier; S2, establishing a linear state equation and a non-linear measurement equation; S3, performing non-linear filtering on a vision-inertia tightly-coupled integrated navigation system by using the PF based on the glowworm swarm optimization to realize data fusion of the vision-inertia tightly-coupled integrated navigation system; and S4, correcting a calculation result of the strapdowninertial navigation system according to a non-linear filtering result obtained in step S3. The vision-inertia tightly-coupled integrated navigation method based on the glowworm swarm optimization PF is suitable for vision-inertia tightly-coupled integrated navigation.

Description

technical field [0001] The invention relates to a visual-inertial tightly coupled combined navigation method, in particular to a visual-inertial tightly coupled combined navigation method based on firefly swarm optimization PF. Background technique [0002] In recent years, individual navigation systems based on various principles have been continuously developed, and their performance has been improved day by day. However, any single navigation sub-equipment or sub-system cannot fully meet the increasing navigation requirements, so the application of integrated navigation technology that can realize complementary advantages is constantly expanding, and has received more and more attention. [0003] The strapdown inertial navigation system has the advantages of low cost, small size, full autonomy, good concealment, and high sampling frequency, but its error diverges with time. The visual odometry based on the visual sensor is a new kind of navigation equipment at present, w...

AI Technical Summary

Problems solved by technology

Particle filter (PF) can optimize important density functions through iterative volumetric Kalman filtering, so as to obtain higher-precision tightly coupled combined filtering results. However, since iterative volumetric Kalman filtering performs volumetric Point calculation and weighted summation, the calculation amount is relatively large, and the real-time performance is poor

At the same time, the particle filter has the problems of particle impoverishment caused by resampling and the need for a large number of particles to perform state estimation.

Images