Search

Loop closure detection is an important component of the Simultaneous Localization and Mapping (SLAM) algorithm, which is utilized in environmental sensing. It helps to reduce drift errors during long-term operation, improving the accuracy and robustness of localization. Such improvements are sorely needed, as conventional visual-based loop detection algorithms are greatly affected by significant changes in viewpoint and lighting conditions. In this paper, we present a semantic spatial structure-based loop detection algorithm. In place of feature points, robust semantic features are used to cope with the variation in the viewpoint. In consideration of the semantic features, which are region-based, we provide a corresponding matching algorithm. Constraints on semantic information and spatial structure are used to determine the existence of loop-back. A multi-stage pipeline framework is proposed to systematically leverage semantic information at different levels, enabling efficient filtering of potential loop closure candidates. To validate the effectiveness of our algorithm, we conducted experiments using the uHumans2 dataset. Our results demonstrate that, even when there are significant changes in viewpoint, the algorithm exhibits superior robustness compared to that of traditional loop detection methods. [ABSTRACT FROM AUTHOR]

In this paper, we propose a data fusion algorithm based on the weighted histogram statistics (DF-WHS) to improve the performance of direction-of-arrival (DOA) estimation for the vector hydrophone vertical array (VHVA). The processing frequency band is firstly divided into multiple sub-bands, and the high-resolution multiple signal classification (MUSIC) algorithm is applied to estimate the azimuth of each sub-band for each vector hydrophone. Then, the weighted least square (WLS) data fusion technique is used to fuse the sub-band estimation results of multiple sensors. Finally, the weighted histogram statistics method is employed to obtain the synthesis results in the frequency domain. We carried out a simulation and sea trial of the 16-element VHVA to evaluate the performance of the proposed algorithm. Compared to several traditional processing algorithms, the beam width of the proposed approach is significantly narrower, the side lobes are considerably lower, and the mean square error (MSE) is effectively smaller. In addition, the DF-WHS method is more suitable to accurately estimate the target azimuth with a low signal-to-noise ratio (SNR) because the noise sub-band is suppressed in the weighted histogram statistics step. The DF-WHS method in this article provides a new approach to improve the performance of deep-sea target detection for the VHVA. [ABSTRACT FROM AUTHOR]