This article investigates the preset performance trajectory tracking control problem of underactuated unmanned surface ships with model uncertainty, unknown external environmental disturbances, and input quantization effects. We consider the non-diagonal damping matrix and mass matrix to satisfy the actual dynamics model of underactuated unmanned surface ships. By adding a hysteresis quantizer, the control method proposed in this article effectively reduces the quantization error. Neural networks are employed to approach the unknown environmental disturbance of underactuated unmanned surface ships. Using the error transformation function, the constrained control problem is transformed into an unconstrained one to ensure the preset performance of tracking errors. This paper verifies the superiority and effectiveness of the proposed control method through Lyapunov stability analysis.

Colonoscopy is a standard imaging tool for examining the lower gastrointestinal tract of patients to capture lesion areas. However, if a lesion area is found during the colonoscopy process, it is difficult to record its location relative to the colon for subsequent therapy or recheck without any reference landmark. Thus, automatic detection of biological anatomical landmarks is highly demanded to improve clinical efficiency. In this article, we propose a novel deep learning-based approach to detect biological anatomical landmarks in colonoscopy videos. First, raw colonoscopy video sequences are pre-processed to reject interference frames. Second, a ResNet-101-based network is used to detect three biological anatomical landmarks separately to obtain the intermediate detection results. Third, to achieve more reliable localization, we propose to post-process the intermediate detection results by identifying the incorrectly predicted frames based on their temporal distribution and reassigning them back to the correct class. Finally, the average detection accuracy reaches 99.75%. Meanwhile, the average intersection over union of 0.91 shows a high degree of similarity between our predicted landmark periods and ground truth. The experimental results demonstrate that our proposed model can accurately detect and localize biological anatomical landmarks from colonoscopy videos.

This paper introduces an innovative staircase shape feature extraction method for walking-aid robots to enhance environmental perception and navigation. We present a robust method for accurate feature extraction of staircases under various conditions, including restricted viewpoints and dynamic movement. Utilizing depth camera-mounted robots, we transform three-dimensional (3D) environmental point cloud into two-dimensional (2D) representations, focusing on identifying both convex and concave corners. Our approach integrates the Random Sample Consensus algorithm with K-Nearest Neighbors (KNN)-augmented Iterative Closest Point (ICP) for efficient point cloud registration. The results show an improvement in trajectory accuracy, with errors within the centimeter range. This work overcomes the limitations of previous approaches and is of great significance for improving the navigation and safety of walking assistive robots, providing new possibilities for enhancing the autonomy and mobility of individuals with physical disabilities.

The accuracy improvement of deep learning-based damage identification methods has always been pursued. To this end, this study proposes a novel damage identification method using Swin Transformer and continuous wavelet transform (CWT). Specifically, the original structural vibration data is first transferred to a time-frequency diagram by CWT, thereby capturing the characteristic information of structural damage. Secondly, the Swin Transformer is applied to learn the two-dimensional time-frequency diagram layer by layer and extract the damage information, by which the damage identification is achieved. Then, the identification accuracy of the proposed method is analyzed under various sample lengths and different levels of environmental noise to validate the robustness of this approach. Finally, the practicality of this method is verified through laboratory test. The results show the proposed method can effectively recognize the damage and achieve excellent accuracy even under noise interference. Its accuracy reaches 99.6% and 99.0% under single damage and multiple damage scenarios, respectively.

Accurate parameter identification of underwater vehicles is of great significance for their controller design and fault diagnosis. Some studies adopt numerical simulation methods to obtain the model parameters of underwater vehicles, but usually only conduct decoupled single-degree-of-freedom steady-state numerical simulations to identify resistance parameters. In this paper, the velocity response is solved by applying a force (or torque) to the underwater vehicle based on the overset grid and Dynamic Fluid-Body Interaction model of STAR-CCM+, solving for the velocity response of an underwater vehicle in all directions in response to propulsive force (or moment) inputs. Based on the data from numerical simulations, a parameter identification method using quantum particle swarm optimization is proposed to simultaneously identify inertia and resistance parameters. By comparing the forward velocity response curves obtained from pool experiments, the identified vehicle model’s mean square error of forward velocity is less than 0.20%, which is superior to the steady-state simulation method and particle swarm optimization and genetic algorithm approaches.