Explore the vast range of titles available.

One of the important stages before the operation of an inertial navigation system is the initial alignment procedure, known as fine alignment under stationary conditions. This study analyzes the accuracy of the fine alignment algorithm, which uses zero-velocity updates to estimate and compensate navigation errors. Since the velocity-matching approach is not fully observable, an observability analysis identifies which error states cannot be estimated. The unobservable states and thus the ultimate alignment accuracy depend on both sensor precision and the orientation of the inertial measurement unit. This paper investigates the influence of attitude on fine alignment performance through analytical derivations and numerical simulations. New relationships are developed to predict alignment accuracy for arbitrary orientations, and their validity is demonstrated via extensive simulation results. Additionally, the impact of x-, y-, and z-channel sensors on the final accuracy of inertial alignment is examined statistically. The results reveal a differential sensitivity, indicating that alignment accuracy is predominantly influenced by sensors along the z-channel. This insight provides a practical strategy for sensor placement optimization, suggesting that allocating higher-accuracy sensors to the z-axis within a fixed budget can yield significant improvements in overall navigation system performance.

Analytical coarse alignment and Kalman filter fine alignment based on zero-velocity are typically used to obtain initial attitude for inertial navigation systems (SINS) on a static base. However, in the shipboard mooring state, the static observation condition is corrupted. This paper presents a rapid alignment method for SINS on swaying bases. The proposed method begins with a coarse alignment technique in the inertial frame to obtain an initial rough attitude. Subsequently, a Kalman filter with position updates is employed to estimate the remaining misalignment error. To enhance the filter estimation performance, an appropriate lower boundary is set to the target states’ variances according to a carefully designed relative convergence index. The variance-constraint Kalman filter (VCKF) approach is proposed in this paper, and the shipborne experiments validate its effectiveness. The results demonstrate that the VCKF approach significantly reduces the time requirement for fine alignment to achieve the same accuracy on a swaying base, from 90 min in the classic Kalman filter to 30 min. Additionally, the parameter estimation performance in the Kalman filter is also improved, particularly in situations where unpredicted external interference is involved during fine alignment.

To solve the problem of slow convergence seen in the traditional fine alignment algorithm based on linear Kalman filtering, a forward–forward backtracking fine alignment algorithm for SINS is proposed after reanalyzing the fine alignment model in this paper. First, the forward–forward backtracking fine alignment model in initial navigation frame was derived. The displacement vector of the carrier in the initial navigation frame solved by GNSS positioning was utilized as the observation of the fine alignment model. Second, under the premise of storing only part of the navigation data, the initial alignment convergence speed was improved by backtracking and reusing the navigation data. The experimental results of the simulation and vehicle tests showed that each backtracking alignment can improve the accuracy of the fine alignment to the performance requirements of the initial alignment, which proved the effectiveness and feasibility of the backtracking fine alignment algorithm proposed in this paper.

Text-to-image person retrieval, a fine-grained cross-modal retrieval problem, aims to search for person images from an image library that match a given textual caption. Existing text-to-image person retrieval methods usually use fixed-point embedding to express the semantics of the two modalities and perform multi-granularity alignment between modalities in the embedding space. However, owing to the inherent mutual one-to-many correspondence between images and texts, it is often difficult for fixed-point embedding methods to adequately capture this relationship, leading to erroneous retrieval results. To address this problem, we propose a novel uncertainty-aware coarse-to-fine alignment method, which first maps fixed-point embedding to probability distributions and then aligns two modalities in terms of distributions and sampling points at a coarse-to-fine granularity, for accurate text-to-image person retrieval. Specifically, we first introduce two contrastive learning tasks of distribution contrast learning and point contrast learning, to achieve coarse-grained inter-modal alignment with uncertainty-aware. The distribution contrast learning task ensures that distributions with the same identity are as similar as possible across modalities through distribution-based contrastive learning. The point contrast learning task performs the contrastive learning of inter-modal and intra-modal sampling points, which not only models rich and diverse cross-modal associations, but also optimizes the learning of distributions. For the fine-grained association requirements of text-to-image person retrieval, we design the task of uncertainty-aware attribute masking language reconstruction, which achieves fine-grained alignment by randomly masking attribute words in the text and reconstructing them via inter-modal sample point interactions. Extensive experiments on two public datasets demonstrate the superior performance of our method.

Precision alignment and target contact are critical tasks for mobile manipulators in industrial inspection and flexible manufacturing. However, achieving high accuracy after navigation remains challenging due to accumulated errors from mobile base localization, perception noise, and calibration uncertainty. This paper proposes a vision-guided precision alignment framework for mobile manipulators using a single front-facing RGB-D camera. The method integrates YOLO-based target detection, AR marker-assisted plane depth estimation, and depth-based 3D localization within a coarse-to-fine alignment strategy. After navigation, the manipulator first moves to a predefined pre-alignment pose, followed by visual localization and iterative refinement to compensate for residual errors before executing precise target contact. The proposed system is implemented and evaluated in a Gazebo-based simulation environment using a mobile manipulator platform model. In a static touch panel experiment with 50 trials, the system achieves a success rate of 98%, with positioning errors maintained within a millimeter-level range. Simulation results demonstrate that the proposed method provides stable alignment performance in the simulation environment without relying on external sensing devices such as force sensors or multi-camera systems. The proposed approach shows promising potential for precision contact tasks in mobile manipulation.

In order to detect the abrasion of rails, a new point cloud alignment method combining 4-points congruent sets (4PCS) coarse alignment based on internal shape signature (ISS) and K-dimensional iterative closest points (KD-ICP) fine alignment is proposed, and for the first time, the combined algorithm is applied to the detection of rail wear. Due to the large amount of 3D rail point cloud data collected by the 3D line laser sensor, the original data are first downsampled by voxel filtering. Then, ISS feature points are extracted from the processed point cloud data for 4PCS coarse alignment, and the feature points are quantitatively analyzed, which in turn provides good alignment conditions for fine alignment. Then, the K-dimensional tree structure is used for the near-neighbor search to improve the alignment efficiency of the ICP algorithm. Finally, the total rail wear is calculated by combining the fine alignment results with the wear calculation formula. The experimental results show that when the number of ISS feature points extracted is 4496, the 4PCS coarse alignment algorithm based on ISS feature points is higher than the original 4PCS algorithm as well as the other algorithms in terms of alignment accuracy; the ICP fine alignment algorithm based on the kd-tree is less than the original ICP algorithm as well as the other algorithms in terms of the time consumed. Further, the proposed new ISS-4PCS + KD-ICP two-stage point cloud alignment method is superior to the original 4PCS + ICP algorithm both in terms of alignment accuracy and runtime. The combined algorithm is applied to the detection of rail wear for the first time, which provides a reference for the non-contact rail wear detection method. The high accuracy and low time consumption of the proposed algorithm lays a good foundation for the calculation of rail wear in the next step.

Cultural and creative products through the visual expression of cultural elements, spirit, symbols, etc., to give the product outside the function of a unique aesthetic experience and spiritual enjoyment, modern three-dimensional printing technology and traditional arts and crafts integration, greatly liberating the traditional fine arts in the shape of the limitations of the designers to facilitate the creative creation. In this paper, the lag in cultural creative products is studied, and the entire process of three-dimensional printing of cultural creative products integrated with the concept of traditional fine arts is designed. Specifically, parametric 3D modeling of cultural and creative products is carried out. Then the dense point cloud model of cultural and creative products is reconstructed by ICP fine alignment. Delaunay triangulation is used to complete the conversion from dense point cloud model to mesh model, and texture mapping is carried out so as to make the generated cultural and creative products more lifelike. The printing of cultural and creative products is completed using SLA technology and printing materials, and their design effectiveness is evaluated by combining with the KANO model. The values of SI and |DSI| are between 31.17%~62.03% and 12.81%~56.02, respectively, and the users’ satisfaction with the cultural and creative 3D printed products integrating the concepts of traditional fine arts is relatively high. Designers can further optimize the 3D printed cultural creative products integrating traditional art concepts on the basis of the original products according to the order of “Basic > Expectation > Charm”.

The navigation means the process of determining the position, velocity, and orientation of the moving object such as the land vehicle, aerial vehicle, and even autonomous vehicle. Nowadays, Global Navigation Satellite System (GNSS) is most used for positioning. Nevertheless, the navigation system based on GNSS would be interrupted in the challenging environments. Therefore, the inertial navigation system (INS) has been widely combined with GNSS to overcome this issue. For INS, the core concept is the measurements of Inertial Measurement Unit (IMU). In order to integrate the measurements of IMU with several sensors such as GNSS receivers, odometers, and so on, we should transform the measurements of IMU from body frame to navigation frame. The initial alignment just represents the process of finding the accurate initial rotation matrix between body frame and navigation frame in the beginning of navigation. However, initial misalignment angles would cause large error of INS. Hence, obtaining an accurate initial rotation matrix from body frame to navigation frame is an important issue to get the better navigation performance. In the research, an integrated navigation system is developed to validate the initial alignment algorithm. With the data pre-processing and accurate calibration. the proposed method of initial alignment can get precise initial ration matrix. The errors of roll pitch, and yaw angle are all smaller than 1 degree after initial alignment. Moreover, the time threshold of initial alignment is set by manual traditionally. A method of finding threshold of coarse alignment is proposed in this research.

Optical and synthetic aperture radar (SAR) image registration faces challenges from nonlinear radiometric distortions and geometric deformations caused by different imaging mechanisms. This paper proposes a coarse-to-fine registration algorithm integrating geographic information constraints with cross-modal feature consistency mapping. The coarse stage employs imaging geometry-based coordinate transformation with airborne navigation data to eliminate scale and rotation differences. The fine stage constructs a multi-scale phase congruency-based feature response aggregation model combined with rotation-invariant descriptors and global-to-local search for sub-pixel alignment. Experiments on integrated airborne optical/SAR datasets demonstrate superior performance with an average RMSE of 2.00 pixels, outperforming both traditional handcrafted methods (3MRS, OS-SIFT, POS-GIFT, GLS-MIFT) and state-of-the-art deep learning approaches (SuperGlue, LoFTR, ReDFeat, SAROptNet) while reducing execution time by 37.0% compared with the best-performing baseline. The proposed coarse registration also serves as an effective preprocessing module that improves SuperGlue’s matching rate by 167% and LoFTR’s by 109%, with a hybrid refinement strategy achieving 1.95 pixels RMSE. The method demonstrates robust performance under challenging conditions, enabling real-time UAV-based multi-sensor fusion applications.

Transfer learning aims to help target learners with a different but related source domain. Open set recognition extends the settings of transfer learning for identifying whether an instance belongs to an unseen category. However, it is often pragmatic and valuable to further classify the unseen categories in target domain. We present a new setting called open set broad classification (OSBC) to classify unseen target categories which are open within the broad classes of source domain. Aiming at adapting to the challenging domain shift between unseen categories and seen categories, we propose a variational autoencoders model with coarse-and-fine alignment (CFVA) to leverage the structural information in the OSBC setting. First, two-stream decoders are employed and coarsely aligned by a relaxed parameters regularizer, which can absorb domain shift on features to facilitate fine alignment. Then fine alignment at encoding level enhances discriminative power of the latent representation by mixing the distributional structure hinted by source domain. Experimental results demonstrate the effectiveness of our CFVA approach in improving the accuracies in both unsupervised and semi-supervised cases.