Sequence-based sparse optimization methods for long-term loop closure detection in visual SLAM

A Robot Pose Estimation Optimized Visual SLAM Algorithm Based on CO-HDC Instance Segmentation Network for Dynamic Scenes

In order to improve the accuracy of visual SLAM algorithms in a dynamic scene, instance segmentation is widely used to eliminate dynamic feature points. However, the existing segmentation technology has low accuracy, especially for the contour of the object, and the amount of calculation of instance segmentation is large, limiting the speed of visual SLAM based on instance segmentation. Therefore, this paper proposes a contour optimization hybrid dilated convolutional neural network (CO-HDC) algorithm, which can perform a lightweight calculation on the basis of improving the accuracy of contour segmentation. Firstly, a hybrid dilated convolutional neural network (HDC) is used to increase the receptive field, which is defined as the size of the region in the input that produces the feature. Secondly, the contour quality evaluation (CQE) algorithm is proposed to enhance the contour, retaining the highest quality contour and solving the problem of distinguishing dynamic feature points from static feature points at the contour. Finally, in order to match the mapping speed of visual SLAM, the Beetle Antennae Search Douglas–Peucker (BAS-DP) algorithm is proposed to lighten the contour extraction. The experimental results have demonstrated that the proposed visual SLAM based on the CO-HDC algorithm performs well in the field of pose estimation and map construction on the TUM dataset. Compared with ORB-SLAM2, the Root Mean Squared Error (Rmse) of the proposed method in absolute trajectory error is about 30 times smaller and is only 0.02 m.

Smart three-dimensional processing of unconstrained cave scans using small unmanned aerial systems and red, green, and blue-depth cameras

This article focuses on a novel three-dimensional reconstruction system that maps large archeological caves using data collected by a small unmanned aircraft system with red, green, and blue-depth cameras. Cave sites often contain the best-preserved material in the archeological record. Yet few sites are fully mapped. Large caves environment usually contains complex geometric structures and objects, which must be scanned with long overlapped camera trajectories for better coverage. Due to the error in camera tracking of such scanning, reconstruction results often contain flaws and mismatches. To solve this problem, we propose a framework for surface loop closure, where loops are detected with a compute unified device architecture accelerated point cloud registration algorithm. After a loop is detected, a novel surface loop filtering method is proposed for robust loop optimization. This loop filtering method is robust to different scan patterns and can cope with tracking failure recovery so that there is more flexibility for unmanned aerial vehicles to fly and record data. We run experiments on public data sets and our cave data set for analysis and robustness tests. Experiments show that our system produces improved results on baseline methods.

Improved and scalable online learning of spatial concepts and language models with mapping

We propose a novel online learning algorithm, called SpCoSLAM 2.0, for spatial concepts and lexical acquisition with high accuracy and scalability. Previously, we proposed SpCoSLAM as an online learning algorithm based on unsupervised Bayesian probabilistic model that integrates multimodal place categorization, lexical acquisition, and SLAM. However, our original algorithm had limited estimation accuracy owing to the influence of the early stages of learning, and increased computational complexity with added training data. Therefore, we introduce techniques such as fixed-lag rejuvenation to reduce the calculation time while maintaining an accuracy higher than that of the original algorithm. The results show that, in terms of estimation accuracy, the proposed algorithm exceeds the original algorithm and is comparable to batch learning. In addition, the calculation time of the proposed algorithm does not depend on the amount of training data and becomes constant for each step of the scalable algorithm. Our approach will contribute to the realization of long-term spatial language interactions between humans and robots.