Abstract:Visual localization plays a critical role in the functionality of low-cost autonomous mobile robots. Contemporary leading methods for precise visual localization are predominantly 3D scene-specific, necessitating extra computational and memory overhead to construct a 3D scene model in novel environments. An alternative approach of directly using a database of 2D images for visual localization offers more flexibility. However, such methods currently suffer from limited localization accuracy. In this paper, we propose an accurate and robust multiple checking-based 3D model-free visual localization system to address the aforementioned issues. To ensure high accuracy, our focus is on estimating the pose of a query image relative to the retrieved database images using 2D-2D feature matches. Theoretically, by incorporating the local planar motion constraint into both the estimation of the essential matrix and the triangulation stages, we reduce the minimum required feature matches for absolute pose estimation, thereby enhancing the robustness of outlier rejection. Additionally, we introduce a multiple-checking mechanism to ensure the correctness of the solution throughout the solving process. The efficacy of our approach is substantiated through both qualitative and quantitative assessments on simulated and two real-world datasets evidencing significant improvements in accuracy and robustness provided by our 3D model-free visual localization system. Note to Practitioners-The motivation of this article stems from the need to develop an accurate visual localization system with simplicity and flexibility of map construction and easy adaption to new environments. Such a system holds great practical value for a range of applications, including warehouse robots, service robots, and countless others. Existing visual localization systems that achieve high accuracy are dependent on a pre-built accurate 3D scene map, which pose challenges in terms of map construction and consume significant storage resources onboard, particularly for large scenes. And the aforementioned efforts need to be repeated when changing to a new scene. In this article, an accurate and robust 3D model-free visual localization system is proposed to handle this problem. The map construction is simplified to build a set of database images with associated camera poses, which is trivial as it amounts to adding posed images to a database. The core idea for achieving high accuracy and robustness is to model the local planar motion characteristic of general ground-moving robots into both essential matrix estimation and triangulation stages to obtain two minimal solutions. The proposed localization system simplifies the task of switching between different application scenarios for the robot, reducing additional workload and lowering the difficulty of use.

Image Jacobian Matrix Estimation Based on Online Support Vector Regression

Visual Servoing with an Uncalibrated Eye-in-hand Camera

3D Model-free Visual Localization System from Essential Matrix under Local Planar Motion

Hybrid Visual Servoing Scheme Applicable for Local Features

Robot visual servoing based on total jacobian

Analyzing Neural Jacobian Methods in Applications of Visual Servoing and Kinematic Control

An Image Based Visual Servo Approach with Deep Learning for Robotic Manipulation

Adaptive Finite-Time Model Estimation and Control for Manipulator Visual Servoing using Sliding Mode Control and Neural Networks

Visual servo of uncalibrated eye-in-hand system with time-delay compensation

Image Gradient-based Joint Direct Visual Odometry for Stereo Camera

Image-Based Composite Learning Robot Visual Servoing With an Uncalibrated Eye-to-Hand Camera

A Motion Planning Method for Visual Servoing Using Deep Reinforcement Learning in Autonomous Robotic Assembly

Robust Kalman filtering with long short-term memory for image-based visual servo control

Image-Based Visual Servoing of Manipulators With Unknown Depth: A Recurrent Neural Network Approach

Dynamic Image-Based Output Feedback Control for Visual Servoing of Multirotors

Image-based Visual Servoing for Mobile Robots

A Digital Twin-Based Visual Servoing with Extreme Learning Machine and Differential Evolution

Robot motion visual measurement based on RANSAC and weighted constraints method

Online robot calibration based on vision measurement

Fully Uncalibrated Image-Based Visual Servoing of 2DOFs Planar Manipulators With a Fixed Camera