Permutation entropy (PE), as one of the powerful complexity measures for analyzing time series, has advantages of easy implementation and high efficiency. In order to improve the performance of PE, some improved PE methods have been proposed through introducing amplitude information and distance information in recent years. Weighted-permutation entropy (W-PE) weight each arrangement pattern by using variance information, which has good robustness and stability in the case of high noise level and can extract complexity information from data with spike feature or abrupt amplitude change. Dispersion entropy (DE) introduces amplitude information by using the normal cumulative distribution function (NCDF); it not only can detect the change of simultaneous frequency and amplitude, but also is superior to the PE method in distinguishing different data sets. Reverse permutation entropy (RPE) is defined as the distance to white noise in the opposite trend with PE and W-PE, which has high stability for time series with varying lengths. To further improve the performance of PE, we propose a new complexity measure for analyzing time series, and term it as reverse dispersion entropy (RDE). RDE takes PE as its theoretical basis and combines the advantages of DE and RPE by introducing amplitude information and distance information. Simulation experiments were carried out on simulated and sensor signals, including mutation signal detection under different parameters, noise robustness testing, stability testing under different signal-to-noise ratios (SNRs), and distinguishing real data for different kinds of ships and faults. The experimental results show, compared with PE, W-PE, RPE, and DE, that RDE has better performance in detecting abrupt signal and noise robustness testing, and has better stability for simulated and sensor signal. Moreover, it also shows higher distinguishing ability than the other four kinds of PE for sensor signals.

In this paper, the Efficient Channel Attention (ECA) mechanism is incorporated at the terminal layer of the YOLOv10 backbone network to enhance the feature expression capability. In addition, Transformer is introduced into the C3 module in the feature extraction process to construct the C3TR module to replace the original C2F module as the deepening network extraction module. In this study, both the ECA mechanism and the self-attention mechanism of Transformer are thoroughly analyzed and integrated into YOLOv10. The C3TR module is used as an important part to deepen the effect of network extraction in backbone network feature extraction. The self-attention mechanism is used to model the long-distance dependency relationship, capture the global contextual information, make up for the limitation of the local sensory field, and enhance the feature expression capability. The ECA module is added to the end of the backbone to globally model the channels of the feature map, distribute channel weights more equitably, and enhance feature expression capability. Extensive experiments on the electrical equipment dataset have demonstrated the high accuracy of the method, with a mAP of 89.4% compared to the original model, representing an improvement of 3.2%. Additionally, the mAP@[0.5, 0.95] reaches 61.8%, which is 5.2% higher than that of the original model.

In order to improve the performance of Autonomous Underwater Vehicles (AUVs) in complex marine environments, efficient and reliable navigation strategies are deemed essential. Over the years, a variety of methods have been developed to solve the underwater vehicle path planning problem. A hybrid path planning method for AUVs operating near the water surface is proposed in this paper. The methodology combines the artificial potential field-based Flow Field Method (FFM) and the Dynamic Window Approach (DWA) to achieve efficient, collision-free navigation. Firstly, a global guidance field is generated by the FFM, which directs the AUV to move toward the goal and avoid static obstacles. The DWA is then integrated to handle real-time local dynamic obstacles, leveraging the AUV’s kinematic constraints. Simulation results demonstrate the effectiveness of this method in complex underwater environment. The adaptability and safety of AUV navigation are significantly improved by this hybrid approach, which is a promising solution in applications of marine exploration, environmental monitoring, and offshore operations.

It is challenging to obtain accurate trajectories with standard camera visual odometry (VO) in environments with weak textures and light variations. This article introduces a novel approach [cubic B-spline-based visual–inertial odometry (CB-VIO)], using the dynamic and active-pixel vision sensor (DAVIS) camera. In the proposed CB-VIO method, the matching mechanism between images and events is designed to improve the success rate of event tracking, based on which the template points from events are utilized to construct a cubic B-spline-based event tracking model within a continuous spatiotemporal window [under SE(3)]. Based on the tracking model to interpolate poses at any time point, the inertial measurement unit (IMU) measurement model is constructed to achieve data fusion from asynchronous and synchronous sensors with different rates. Compared with the Spline-visual–inertial odometry (VIO) and the event-based VO (EVO), the proposed continuous spatiotemporal window method can effectively solve the data association for EVO and the continuous-time trajectory with fixed-time intervals for Spline-VIO. The experimental results are compared on public datasets of multiple different scenes, which demonstrate the superior performance of CB-VIO in terms of accuracy and robustness (translation error ${\leq }1.3\%$ and rotation error ${\leq }2^{\circ} $ ).

As a new type of vision sensor, the dynamic and active-pixel vision sensor (DAVIS) outputs image intensity and asynchronous event streams in the same pixel array. We present a novel visual odometry algorithm based on the DAVIS in this paper. The Harris detector and the Canny detector are utilized to extract an initialized tracking template from the image sequence. The spatio-temporal window is selected by determining the life cycle of the asynchronous event streams. The alignment on timestamps is achieved by tracking the motion relationship between the template and events within the window. A contrast maximization algorithm is adopted for the estimation of the optical flow. The IMU data are used to calibrate the position of the templates during the update process that is exploited to estimate camera trajectories via the ICP algorithm. In the end, the proposed visual odometry algorithm is evaluated in several public object tracking scenarios and compared with several other algorithms. The tracking results show that our visual odometry algorithm can achieve better accuracy and lower latency tracking trajectory than other methods.

In complex indoor environments—such as chaotic warehouses or post-disaster sites—achieving accurate simultaneous localization and mapping (SLAM) is hindered by inherent limitations of the sensors involved: Light Detection and Ranging (LiDAR) loses effectiveness in glass corridors, and Red–Green–Blue-Depth (RGB-D) cameras suffer from performance degradation when exposed to poor lighting conditions. To address this, we proposes a real-time multi-sensor fusion SLAM framework that integrates 2D LiDAR and RGB-D cameras. An improved particle filter–based IP-Mapping algorithm is introduced, redesigning the proposal distribution and incorporating KLD adaptive resampling, which accelerates particle convergence by 1.5 × compared with Gmapping while maintaining mapping accuracy. Additionally, a novel I-SURF feature matching method combines ORB with edge-enhanced descriptors, achieving 96.8% matching accuracy under varying illumination and scale conditions. A region-based fusion algorithm is further developed to align RGB-D point clouds with 2D LiDAR scans using octree and KD-tree structures. Experiments in office-like environments demonstrate that the proposed framework reduces localization error by 75.4%–83.5% relative to single-sensor baselines, achieving a grid map resolution of 0.05 m/cell in real time. These results validate the effectiveness of the method for robust robotic mapping and navigation in cluttered indoor settings.