Abstract:Motivated by the goal of enhancing the accuracy and robustness of visual inertial navigation systems(VINSs) across a wide spectrum of dynamic scenarios, protracted missions and expansive navigation ranges, we designed a monocular visual inertial odometry (VIO) augmented by planar environmental constraints. To attain efficient feature extraction and precise feature tracking, we employed a methodology that involved the extraction and tracking of uniformly distributed using features from accelerated segment test(FAST) feature points from video images, with the subsequent removal of outliers through symmetric optical flow. Additionally, we outlined the process of identifying coplanar feature points from the sparse feature set, enabling efficient plane detection and fitting. This approach constructed spatial geometric constraints on the three-dimensional coordinates of visual feature points without resorting to computationally expensive dense depth mapping. The heart of this method lied in the formulation of a comprehensive cost function, which integrated the reprojection error of visual feature points, the coordinate constraints derived from coplanar feature points, and the inertial measurement unit (IMU) pre-integration error. These integrated measurements were then utilized to estimate the system states through a nonlinear optimization methodology. To validate the accuracy and effectiveness of the proposed approach, extensive experiments were conducted using publicly available datasets and large-scale outdoor scenes. The experimental results conclusively demonstrate that compared to VINS-Mono and ORB-SLAM3, the proposed method achieves higher positioning accuracy. It can deliver precise and stable navigation results even in challenging conditions, thereby imparting significant practical value to the fields of robotics and unmanned driving.

Monocular Visual-Inertial Odometry Based on Sparse Feature Selection with Adaptive Grid

Motion estimation based feature selection for visual SLAM

Visual Odometry Research Based on Efficient Feature Tracking

Direct Sparse Monocular Visual-Inertial Odometry with Covisibility Constraints

Guided Feature Selection for Deep Visual Odometry

Rapid and Robust Monocular Visual-Inertial Initialization with Gravity Estimation Via Vertical Edges.

Balancing the Budget: Feature Selection and Tracking for Multi-Camera Visual-Inertial Odometry

Hierarchical Quadtree Feature Optical Flow Tracking Based Sparse Pose-Graph Visual-Inertial SLAM

Robust and Efficient Visual-Inertial Odometry with Multi-plane Priors.

Fast and Robust Monocular Visua-Inertial Odometry Using Points and Lines

Angular Tracking Consistency Guided Fast Feature Association for Visual-Inertial SLAM

A Dense Optical Flow-Based Feature Matching Approach in Visual Odometry

Visual Odometry Based on Improved Feature Matching and Unscented Kalman Filter

Fast Visual Inertial Odometry with Point–line Features Using Adaptive EDLines Algorithm

A Robust and Integrated Visual Odometry Framework Exploiting the Optical Flow and Feature Point Method

Planar environmental constraints aided monocular visual inertial odometry

VI-HSO: Hybrid Sparse Monocular Visual-Inertial Odometry

Fast Sparse LiDAR Odometry Using Self-Supervised Feature Selection on Intensity Images

Robust Visual Inertial Monocular Using Nonlinear Optimization.

Tightly Coupled Visual-Inertial Fusion with Image Enhancement for Robust Positioning

Fast and Accurate Visual Odometry from a Monocular Camera