Robot SLAM method and system used in outdoor feature sparse environment

Abstract

The invention provides a mobile robot vision inertia SLAM method and system used in an outdoor feature sparse environment. The method comprises the steps: obtaining an environment image in the outdoor feature sparse environment, carrying out the preprocessing of the environment image, and obtaining a preprocessed environment image; extracting sparse features of the preprocessed environment image through a block SIFT feature extraction algorithm; estimating the inter-frame motion of a robot through the pre-integration quantity of an inertial unit IMU, estimating the inter-frame displacement of a matching point, judging the importance of sparse features through the displacement, and calculating to obtain a visual re-projection error; obtaining an IMU error according to variance accumulation in an IMU pre-integration process; jointly constructing a loss function according to the visual re-projection error and the IMU error; and minimizing a loss function through a nonlinear optimization method, and solving pose transformation of the robot and space coordinates of map points. According to the invention, the influence of difficulties such as a violent illumination change and sparse features on the positioning performance can be reduced, and the autonomy of the mobile robot in a complex outdoor environment is improved.

Description

technical field [0001] The present invention relates to a robot real-time positioning and mapping algorithm, in particular, to a robot SLAM method and system used in an outdoor environment with sparse features, and more specifically, to a robot visual inertia system used in an outdoor environment with sparse features Real-time positioning and mapping algorithms, using image preprocessing, feature detection, and optimizing visual error loss weights to improve the positioning and mapping capabilities of robots in outdoor scenes with sparse features. Background technique [0002] The general SLAM problem is aimed at scenes with relatively closed environments such as indoors and streets and rich environmental features. In environments with sparse outdoor features (such as deserts, sandy beaches, rocky beaches, planets and lunar surfaces, etc.), the problem of real-time positioning and mapping of robots is related to the autonomous positioning, safety, and mission planning of rob...

AI Technical Summary

Problems solved by technology

[0003] The publication number is CN110309834A (application number: 201910393343.1), which discloses a SLAM algorithm for outdoor offline navigation. A certain number of feature points are obtained through a locally adaptive threshold adjustment algorithm, which overcomes the shadows and weak lighting in the images of wild scenes. The problem of insufficient accuracy of feature point extraction under the conditions of objective factors such as sudden noise and sudden noise

This method does not fundamentally solve the problems of difficult feature extraction and drastic illumination changes in the outdoor environment with sparse features. For the sparse environment, lowering the feature detection threshold will bring feature instability, which will cause greater errors; and use multi-scale detection. The method of the method is relatively vague for the description of visual features, and cannot accurately measure the weight of image features in optimization.

Images