Zheng Rong

Pose graph optimization helps reduce drift accumulated in pure odometry of visual simultaneous localization and mapping (SLAM) systems by solving a nonlinear least square problem, including both sequential constraints and loop-closing constraints. However, the covariances of all constraints are set to constant matrices or by manual setting. In this paper, we propose a novel approach to approximate covariances of constraints in pose graph optimization to better represent the true uncertainty of the underlying visual-inertial navigation system (VINS) that fuses inertial measurements and visual observations. Specifically, for sequential constraints, we propose to utilize nonlinear factor recovery to optimally extract covariance matrices from the accumulated visual-inertial odometry (VIO). For loop-closing constraints, we propose a dynamic scale estimation method to approximate the scales of the information matrices. To evaluate the effectiveness and robustness of the proposed method, we conduct extensive experiments on public and self-collected datasets in various environments. Results show that our proposed method achieves higher accuracy compared with naively-formulated pose graph optimization adopted by several state-of-the-art visual-inertial navigation systems.

Most image matching methods perform poorly when encountering large scale changes in images. To solve this problem, we propose a Scale-Difference-Aware Image Matching method (SDAIM) that reduces image scale differences before local feature extraction, via resizing both images of an image pair according to an estimated scale ratio. In order to accurately estimate the scale ratio for the proposed SDAIM, we propose a Covisibility-Attention-Reinforced Matching module (CVARM) and then design a novel neural network, termed as Scale-Net, based on CVARM. The proposed CVARM can lay more stress on covisible areas within the image pair and suppress the distraction from those areas visible in only one image. Quantitative and qualitative experiments confirm that the proposed Scale-Net has higher scale ratio estimation accuracy and much better generalization ability compared with all the existing scale ratio estimation methods. Further experiments on image matching and relative pose estimation tasks demonstrate that our SDAIM and Scale-Net are able to greatly boost the performance of representative local features and state-of-the-art local feature matching methods.

Simultaneous localization and mapping (SLAM) has been studied for decades in the field of robotics, in which light detection and ranging (LiDAR) is widely used in various application areas benefiting from its accessibility of direct, accurate, and reliable 3-D measurements. However, the performance of LiDAR SLAM may be degraded when running in degenerate scenario, which makes it still a challenging problem to realize real-time, robust, and accurate state estimation in complex environments. In this article, we propose a keyframe-based 3-D LiDAR SLAM using an accurate principal component analysis (PCA)-based feature extraction method and an efficient two-stage matching strategy, toward a more robust, accurate, and globally consistent estimation performance. The effectiveness and performance are demonstrated and evaluated by comparing our method with the state-of-the-art open-source methods, LOAM and LeGo-LOAM, on KITTI datasets and custom datasets collected by our sensor system. The experimental results show obvious improvement of odometry accuracy and mapping consistency without loss of real-time performance.

Accurate localization and reliable mapping is essential for autonomous navigation of robots. As one of the core technologies for autonomous navigation, Simultaneous Localization and Mapping (SLAM) has attracted widespread attention in recent decades. Based on vision or LiDAR sensors, great efforts have been devoted to achieving real-time SLAM that can support a robot’s state estimation. However, most of the mature SLAM methods generally work under the assumption that the environment is static, while in dynamic environments they will yield degenerate performance or even fail. In this paper, first we quantitatively evaluate the performance of the state-of-the-art LiDAR-based SLAMs taking into account different pattens of moving objects in the environment. Through semi-physical simulation, we observed that the shape, size, and distribution of moving objects all can impact the performance of SLAM significantly, and obtained instructive investigation results by quantitative comparison between LOAM and LeGO-LOAM. Secondly, based on the above investigation, a novel approach named EMO to eliminating the moving objects for SLAM fusing LiDAR and mmW-radar is proposed, towards improving the accuracy and robustness of state estimation. The method fully uses the advantages of different characteristics of two sensors to realize the fusion of sensor information with two different resolutions. The moving objects can be efficiently detected based on Doppler effect by radar, accurately segmented and localized by LiDAR, then filtered out from the point clouds through data association and accurate synchronized in time and space. Finally, the point clouds representing the static environment are used as the input of SLAM. The proposed approach is evaluated through experiments using both semi-physical simulation and real-world datasets. The results demonstrate the effectiveness of the method at improving SLAM performance in accuracy (decrease by 30% at least in absolute position error) and robustness in dynamic environments.

Robust and accurate localization and mapping are essential for autonomous driving. The traditional SLAM methods generally work under the assumption that the environment is static, while in dynamic environment the performance will be degenerate. In this paper, we propose an efficient and effective method to eliminate the influence of dynamic environment on SLAM by fusing LiDAR and mmW-radar, which significantly improves the robustness and accuracy of localization and mapping. The method fully utilizes the advantages of different measurement characteristics of two sensors, efficient moving object detection based on Doppler effect by radar and accurate object segmentation and localization by LiDAR, to remove the moving objects and uses the resulting filtered point cloud as the input of SLAM towards enhanced performance. The proposed approach is evaluated through experiments in various real world scenarios, and the results demonstrate the effectiveness of the method to improve the robustness and accuracy of SLAM in dynamic environments.

We pursue the formulation of an online methodology to detect future motions that can result in degraded perceptual information leading to state estimation inaccuracy and degeneracy. As autonomous systems and first responders engage in an unknown environment, real-time perceptual and state estimation systems enable accurate localization with respect to the environment. We propose a methodology leveraging numerical techniques in nonlinear observability analysis to enable online detection of degenerate conditions and the computation of motions leading to degradation in localization performance. The approach is specialized to a representative monocular visual-inertial state estimation formulation and evaluated through simulation and experiments to assess the performance of the system to both detect degenerate conditions and identify motions that may lead to future degradation or the failure of the localization system.