Graph Optimization Research Articles

An Improved Inertial Preintegration Model in Factor Graph Optimization for High Accuracy Positioning of Intelligent Vehicles

Accurate and reliable positioning is of great importance for the realization of intelligent vehicles (IV). Factor graph optimization (FGO) has been popularized in the field of robotics for state estimation. As a ubiquitous sensor, the IMU is widely used for vehicular positioning based on the preintegration theory using FGO. However, the existing preintegration model fails to consider the earth's rotation, which challenges the attitude and positioning performance of vehicles equipped with high-precision inertial measurement units (IMU). To fill this gap, an accurate IMU factor that accounts for the earth's rotation rate is designed in this paper. First, an exact preintegration measurement model is derived based on a high-precision inertial navigation system (INS) mechanization. The proposed factor is more accurate and it enables the application of the FGO more suitable for inertial-based integrated navigation with different precision IMUs. In both simulation and field tests, the INS/ Global Navigation Satellite System (GNSS) integration, as a representative of inertial-based integrated navigation, is used to verify the performance of the proposed preintegration factor. The navigation results using our experimental car reveal that the proposed model leads to a more accurate estimation, outperforming the traditional preintegration models, especially in the aspect of attitude estimation.

3D Reconstruction with Spherical Cameras

The goal of image-based 3D reconstruction is to establish a high-quality 3D expression from images. In order to achieve a high-resolution and real-time 3D model, inspired by the open source COLMAP, we propose a novel framework (3DMAP) to reconstruct 3D scenes based on spherical cameras. Unlike traditional methods which focus on building a 3D plane by Poisson distribution function, our method illustrates the key processes of 3D reconstruction: locating the camera based on global feature, estimating the scene’s relative depth from monocular panoramic images, and obtaining a high-quality 3D surface reconstruction. In the camera locating part, we use a global descriptor augmentation model to build a labeled panorama dataset GDAP, in which the images are captured by our designed spherical cameras; In the depth estimation part, we propose a new network UMDE that can estimate the depth of both indoor and outdoor scenes; Finally, in the 3D surface reconstruction section, we turn the reconstruction problem to a graph optimization problem, called GraphFit, in which, we optimize the point clouds with s-t graph and smoothing method successively. We conduct experiments on our own dataset to demonstrate the effectiveness of our proposed 3DMAP framework. Experimental results show that our 3DMAP has achieved good evaluation scores and visual effects.

BayGO: Decentralized Bayesian Learning and Information-Aware Graph Optimization Framework

BayGO: Decentralized <b>Bay</b>esian Learning and Information-Aware <b>G</b>raph <b>O</b>ptimization Framework

A Robust Factor Graph Optimization Method for Navigation in Land Vehicles Based on Dynamic Kernel Principal Component Analysis

A Robust Factor Graph Optimization Method for Navigation in Land Vehicles Based on Dynamic Kernel Principal Component Analysis

Lidar SLAM Method for Photovoltaic Power Stations Based on Factor Graph Optimization

Lidar SLAM Method for Photovoltaic Power Stations Based on Factor Graph Optimization

Construction of Probability Graph Optimization Model on Basis of Localization Algorithm

Accurate positioning is crucial in fields such as indoor positioning, autonomous driving, and robot navigation. However, traditional base station positioning algorithms are often affected by factors such as sensor errors, environmental noise, and uncertainty, resulting in certain errors in the positioning results. The article constructed a probability graph model using Radio Frequency Identification (RFID) by using position estimation results as variables. In this model, each position estimation result was represented as a node, and the dependency relationship between positions was represented as an edge. This article observed sensor data and uses Bayesian inference methods to update the probability distribution of each node. The experimental results showed that compared to traditional base station positioning algorithms, the probability graph optimization model based on radio frequency identification can significantly improve the accuracy and reliability of positioning. The highest values for both reached 0.989 and 0.95, respectively. This model has significant advantages in positioning tasks and provides an effective solution for practical applications.

Vehicle Heading Enhancement Based on Adaptive Sliding Window Factor Graph Optimization for Gyroscope/Magnetometer

Vehicle Heading Enhancement Based on Adaptive Sliding Window Factor Graph Optimization for Gyroscope/Magnetometer

Robust pose graph optimization against outliers using consistency credibility factor

Robust pose graph optimization against outliers using consistency credibility factor

LiDAR Inertial SLAM Algorithm Based on IESKF with Factor Graph Optimization

LiDAR Inertial SLAM Algorithm Based on IESKF with Factor Graph Optimization

Plane Constraints Aided Multi-Vehicle Cooperative Positioning Using Factor Graph Optimization

Plane Constraints Aided Multi-Vehicle Cooperative Positioning Using Factor Graph Optimization

Neural Network Aided Factor Graph Optimization for Collaborative Pedestrian Navigation

Neural Network Aided Factor Graph Optimization for Collaborative Pedestrian Navigation

Hierarchical Distribution-Based Tightly-Coupled LiDAR Inertial Odometry

LiDAR inertial odometry (LIO) has attracted much attention due to the complementarity of LiDAR and IMU measurements. In the distribution-based LIO, the components related to distribution covariance in the residual and residual uncertainty from the LiDAR measurement noise is neutralized. And the resultant point cloud constraint degeneration problem severely affects the accuracy of pose estimation. In this paper, a hierarchical tightly-coupled LIO based on distribution is proposed. By excluding the eigenvalue elements in the distribution covariance component with the designed loss function, the uncertainty of corresponding residual is rectified. As a result, the degeneration problem is solved. With anti-degeneration point-to-distribution constraints, a LiDAR inertial odometry based on iterated extended Kalman filter and a factor graph optimization are designed and organized in a hierarchical way to achieve coarse-to-fine pose estimation, where LiDAR and IMU measurements are tightly coupled in both layers. In this way, the respective advantages of high efficiency and high accuracy from filtering and optimization are combined, which offers high-fidelity estimation results in real time. The effectiveness of the proposed method is verified through experiments on the public NC and ENC datasets.

Majorization Minimization Methods for Distributed Pose Graph Optimization

Majorization Minimization Methods for Distributed Pose Graph Optimization

Time-Differenced Carrier Phase Assisted Visual-Inertial Odometry with Global Pose Graph Optimization

Time-Differenced Carrier Phase Assisted Visual-Inertial Odometry with Global Pose Graph Optimization

Integrity-Constrained Factor Graph Optimization for GNSS Positioning in Urban Canyons

Integrity-Constrained Factor Graph Optimization for GNSS Positioning in Urban Canyons

Enhancing Smartphone-based Pedestrian Positioning: Using Factor Graph Optimization with Indoor/Outdoor Detection for 3DMA GNSS/Visual-Inertial State Estimation

This paper explores the pervasive challenges of pedestrian positioning using smartphones in densely populated urban environments where Global Navigation Satellite System (GNSS) signals are inaccessible, for example, in indoor areas. Existing sensor-based positioning methods, such as inertial navigation systems (INS), GNSS, and visual-inertial odometry (VIO), suffer from inherent restrictions that compromise the accuracy and reliability of the positioning performance. An approach based on machine learning is proposed to address these limitations, employing the Support Vector Machine (SVM) algorithm to accurately distinguish indoor/outdoor (IO) based on the measurement of GNSS. The proposed approach in this study seamlessly incorporates 3D mapping aided (3DMA) GNSS measurements and localized estimations derived by VIO via factor graph optimization (FGO), complemented by an IO detection switch, to achieve accurate pose estimation and effectively eliminate global drift. The system's effectiveness and robustness are rigorously assessed through comprehensive extensive real-life experiments, with an average reduction of 4 meters, leading to noteworthy and statistically significant findings.

Windowing-Based Factor Graph Optimization With Anomaly Detection Using Mahalanobis Distance for Underwater INS/DVL/USBL Integration

Windowing-Based Factor Graph Optimization With Anomaly Detection Using Mahalanobis Distance for Underwater INS/DVL/USBL Integration

Intelligent Environment-Adaptive GNSS/INS Integrated Positioning with Factor Graph Optimization

Global navigation satellite systems (GNSSs) applied to intelligent transport systems in urban areas suffer from multipath and non-line-of-sight (NLOS) effects due to the signal reflections from high-rise buildings, which seriously degrade the accuracy and reliability of vehicles in real-time applications. Accordingly, the integration between GNSS and inertial navigation systems (INSs) could be utilized to improve positioning performance. However, the fixed GNSS solution uncertainty of the conventional integration method cannot determine the fluctuating GNSS reliability in fast-changing urban environments. This weakness becomes solvable using a deep learning model for sensing the ambient environment intelligently, and it can be further mitigated using factor graph optimization (FGO), which is capable of generating robust solutions based on historical data. This paper mainly develops the adaptive GNSS/INS loosely coupled system on FGO, along with the fixed-gain Kalman filter (KF) and adaptive KF (AKF) being taken as comparisons. The adaptation is aided by a convolutional neural network (CNN), and the feasibility is verified using data from different grades of receivers. Compared with the integration using fixed-gain KF, the proposed adaptive FGO (AFGO) maintains the 100% positioning availability and reduces the overall 2D positioning error by up to 70% in the aspects of both root mean square error (RMSE) and standard deviation (STD).

An Enhanced Multi-Sensor Simultaneous Localization and Mapping (SLAM) Framework with Coarse-to-Fine Loop Closure Detection Based on a Tightly Coupled Error State Iterative Kalman Filter

In order to attain precise and robust transformation estimation in simultaneous localization and mapping (SLAM) tasks, the integration of multiple sensors has demonstrated effectiveness and significant potential in robotics applications. Our work emerges as a rapid tightly coupled LIDAR-inertial-visual SLAM system, comprising three tightly coupled components: the LIO module, the VIO module, and the loop closure detection module. The LIO module directly constructs raw scanning point increments into a point cloud map for matching. The VIO component performs image alignment by aligning the observed points and the loop closure detection module imparts real-time cumulative error correction through factor graph optimization using the iSAM2 optimizer. The three components are integrated via an error state iterative Kalman filter (ESIKF). To alleviate computational efforts in loop closure detection, a coarse-to-fine point cloud matching approach is employed, leverging Quatro for deriving a priori state for keyframe point clouds and NanoGICP for detailed transformation computation. Experimental evaluations conducted on both open and private datasets substantiate the superior performance of the proposed method compared to similar approaches. The results indicate the adaptability of this method to various challenging situations.

Fast monocular visual-inertial calibration and initialization with weak environmental control and motion control

Most existing visual-inertial systems require control of extrinsic environment and extrinsic motion to obtain calibration results and simplify the initialization model. The former limits the usage scenarios of the method, while the latter reduces the estimation accuracy of the state parameters of the navigation system. Therefore, a visual-inertial calibration and initialization method is necessary for weak environmental and motion control conditions. Firstly, to overcome the reliance on environmental control, we establish a motion constraint equation based on the relative consistency of visual-inertial sensing spatial motion and implement a visual-inertial online calibration method. Secondly, to eliminate the reliance on motion control, a dual graph optimization structure is be constructed. The initial optimization outcome serves as the foundation for achieving improved calibration results in the refined optimization. Thirdly, to improve initialization accuracy, a complete initial state model is considered to decouple and calculate the translation extrinsic parameters. These parameters are compensated into subsequent initialization, enhancing the robustness of the initialization process. The proposed method is validated on the EuRoC dataset, and experimental results demonstrate that it achieves more accurate calibration results at a faster rate than VINS-Mono in scenarios without calibration target and with limited vehicle motion. The navigation results of different scenarios demonstrate that our method can more effectively integrate visual-inertial data due to improved calibration and initialization estimates.