Abstract:With the popularization of consumer-grade low-cost robots, visual SLAM technology has been widely used for robot localization and navigation in indoor and outdoor environments. However, most current visual SLAM systems suffer from poor accuracy and robustness of pose estimation due to their vulnerability to feature extraction challenges, mismatching, and tracking loss in illumination changes and low-texture environments. Aiming to reduce the localization error of the robot and improve its operational stability, this paper proposes a novel stereo vision SLAM system based on deep neural networks. Firstly, in view of the fact that the handcrafted visual features have problems such as unstable feature extraction and inaccurate feature matching in environments with illumination changes and low texture, we designed a visual feature extraction module based on feature extraction network to overcome these issues, which contains a differentiable keypoint detection module, as well as reprojection and dispersed peak losses for accurate and repeatable keypoint training. Secondly, since the feature similarity measure of the bag-of-word algorithm uses Hamming distance, it cannot support the query and matching of visual features of deep neural networks. We improved it and generated a new visual feature dictionary based on deep neural network extraction to ensure the correctness of visual feature query and matching. Finally, in order to further optimize the pose of SLAM system, based on the original vision odometry, we use the redundant pose information of Lidar to complementary fuse the pose information of the visual front end and the laser front end. We apply the algorithm to the KITTI dataset and an actual mobile robot platform. Experimental results show that compared with other methods, the approach proposed in this article effectively enhances the accuracy and robustness of the SLAM system, and also proves the practicability and effectiveness of the method.

Real-Time and Uniform Feature Filtering for Stereo Images Based on Spatial Indexing

Stereo Vision Based SLAM Using Rao-Blackwellised Particle Filter

Performance evaluation of feature detection and matching in stereo visual odometry.

Stereo Vision Based SLAM in Outdoor Environments

A robust stereo feature-aided semi-direct SLAM system

Full-Image Guided Filtering for Fast Stereo Matching

A Real-time Multi-lane Detection Approach Using Features Geometric Uncertainty and Stereo Vision

Visual Odometry Research Based on Efficient Feature Tracking

Edge-preserving Guided Filtering Based Cost Aggregation for Stereo Matching.

Stereo matching algorithm using improved guided filtering

Accurate Real-Time Stereo Correspondence Using Intra- and Inter-Scanline Optimization

An improved point feature-based sparse stereo vision

Stereo Matching Algorithm for Robot Binocular Images in Real Time

Long-Range Outdoor Localization Of A Mobile Robot Using A Binocular Camera

A Real-time Stereo Visual-Inertial SLAM System Based on Point-and-Line Features

Noise Robust Image Matching Using Adjacent Evaluation Census Transform and Wavelet Edge Joint Bilateral Filter in Stereo Vision

Real-Time Stereo 3D Car Detection With Shape-Aware Non-Uniform Sampling

Robust stereo inertial odometry based on self-supervised feature points

Stereo Vision SLAM Based on Feature Extraction Network

Robust and accurate depth estimation by fusing LiDAR and Stereo

Efficient Stereo Matching with Adaptive Disparity Range Scaling and Feature Calibration