The ability to achieve controllable multimodal locomotion on complex terrains is crucial for the practical applications of small-scale legged robots. In this study, a novel magnetically actuated soft quadrupedal terrestrial millirobot was designed. Inspired by biological terrestrial locomotion modes, three distinct locomotion modes-quadrupedal bounding, quadrupedal pacing, and bipedal walking-were realized through a combination of various postures under a uniform external magnetic field and asymmetrical friction effects induced by magnetic torque. The characteristics of these modes were examined and compared, including the effects of magnetic field strength, swing angle, and surface roughness on stride length. Furthermore, the line-of-sight control method was implemented in path-tracking experiments, enabling closed-loop control on complex paths and improving tracking accuracy. This research holds significant potential for applying magnetically controlled small-scale robots in the bioengineering and industrial micromanipulation fields.

A multi-feature semi-parametric modeling method is proposed aiming at the inverse dynamics modeling problem of the 7-DOF heavy-duty hydraulic manipulator (HDHM), which is widely applied in construction machinery. The method can compensate for the influence of nonlinear factors on manipulator inverse dynamics modeling, and further improve the accuracy and generalization ability of the dynamics model. First, the construction process of the parametric model is de-scribed in detail. Then, based on recurrent neural network (RNN), a multi-feature semi-parametric dynamics model (MSDM) is constructed and validated under two operating conditions (high-speed and medium-load, high-speed and heavy-load) of the 7-DOF HDHM. The experimental results show that the MSDM model performs optimally compared to the other models mentioned in this paper, laying the foundation for path optimization and control of the 7-DOF HDHM.

In this paper, a new solution is proposed for the problem of mooring safety of large ships in complex sea conditions. Firstly, a dual-mode mooring system is designed to adaptively switch between active control and passive energy storage, adjusting the mooring strategy based on real-time sea conditions. Second, a collaborative analysis platform based on AQWA-Python-MATLAB/Simulink was researched and developed. Thirdly, based on the above simulation platform, the performance of the mooring system and the effects of different configurations on the stability of ship motion and dynamic tension of the cable are emphasized. Finally, by comparing the different mooring positions under various sea conditions with the traditional mooring system, the results show that the constant tension mooring system significantly improves the stability and safety of the ship under both conventional and extreme sea conditions, effectively reducing the fluctuation of cable tension. Through the optimization analysis, it is determined that the configuration of bow and stern cables is the optimal solution, which ensures safety while also improving economic benefits.

Magnetic miniature swimmers hold significant potential for applications in biomedical and other fields. However, their precise control remains a major challenge due to fabrication imperfections and magnetic field in- homogeneities. In this paper, we designed a small-size swimmer composed of magnetoelastic composite materials, capable of swimming on liquid surfaces. To address the deviation between the desired angle and the actual swimmer angle, and to improve motion control accuracy, an inverse modeling approach with deep neural networks (IMDNN) was proposed for error compensation. In open-loop experiments, IMDNN compensation reduced angle error and position error by 75.2 % and 85.9 %, respectively. When integrated into a closed-loop control framework combining the line-of-sight (LOS) navigation method and PID control, IMDNN further enhanced trajectory tracking precision and disturbance rejection. Closed-loop path tracking experiments demonstrated reductions of 67.7 % and 77.6 % in angle error and position error, respectively, compared to control without compensation. The results show that the proposed error compensation method significantly reduces path tracking errors, providing a novel strategy for the precise control of miniature swimmers. This method offers a promising solution for applications in biomedical micromanipulation and targeted drug delivery fields.

This paper proposes a shore-based constant tension mooring system, which improves the cable tension distribution by adjusting the length of the cable to maintain the constant tension of the cable between the ship and the mooring pile in order to solve the problem of poor safety and reliability of the traditional mooring system in the mooring process. First, based on the three-dimensional potential flow theory, this paper uses the hydrodynamic software AQWA to numerically simulate the dynamic response of the traditional mooring system under the coupling of wind, wave, and current in different sea states. Subsequently, a shore-based constant tension mooring system using the principle of volume-varying hydraulic control was studied. On the basis of a comprehensive analysis of the working principle of the constant tension hydraulic control mooring system, a mathematical model of the main working circuit is established. The system was numerically simulated by relying on matlab/Simulink simulation software. Finally, by comparing with traditional mooring systems, the results show that the maximum cable tension of the shore-based constant tension mooring system is significantly reduced so that the tension is controlled within a fixed range, and the safety factor of the mooring cable is significantly improved, thus reducing the risk of mooring system failure and improving the ship's survivability.

The task of instance segmentation is widely acknowledged as being one of the most formidable challenges in the field of computer vision. Current methods have low utilization of boundary information, especially in dense scenes with occlusion and complex shapes of object instances, the boundary information may become ineffective. This results in coarse object boundary masks that fail to cover the entire object. To address this challenge, we are introducing a novel method called boundary-guided global feature fusion (BGF) which is based on the Mask R-CNN network. We designed a boundary branch that includes a Boundary Feature Extractor (BFE) module to extract object boundary features at different stages. Additionally, we constructed a binary image dataset containing instance boundaries for training the boundary branch. We also trained the boundary branch separately using a dedicated dataset before training the entire network. We then input the Mask R-CNN features and boundary features into a feature fusion module where the boundary features provide shape information needed for detection and segmentation. Finally, we use a global attention module (GAM) to further fuse features. Through extensive experiments, we demonstrate that our approach outperforms state-of-the-art instance segmentation algorithms, producing finer and more complete instance masks while also improving model capability.

The armor layer unit of the breakwater is a commonly used structure in ocean engineering. They are usually densely arranged and often have complex occlusion and overlap. These factors make traditional segmentation methods difficult to meet the requirements of high precision and efficiency. Unlike existing approaches to generic target segmentation, we propose a specific target segmentation method characterized by segmenting small and dense instances with similar characteristics. Our approach called the attention-guided enhanced feature pyramid network (AE-FPN) for breakwater armor layer unit segmentation, consists of two major components. The first component is an attention-guided (AM) module. Comprising both a channel context attention module (CCAM) and a spatial context attention module (SCAM), the AM module uses contextual information to allow the model to learn information about the region containing the target. The second component is semantic feature enhancement (SFE), wherein pyramid-like structures are employed to enhance semantic information. To assess the performance of the AE-FPN, a task-specific breakwater armor unit dataset named SUD2022 was released. Without any bells and whistles, the proposed AE-FPN achieved 75.2% AP on this dataset, which represents a 9.1% improvement over that of the Mask R-CNN. We also performed ablation experiments on the Cifar10 and COCO datasets to verify the generalization of our designed module. Large-scale experimental results on SUD2022 and COCO datasets demonstrate that the AE-FPN not only achieves excellent performance on breakwater armor layer unit segmentation but can also improve the performance of generic object segmentation.

针对七自由度重载液压机械臂,建立精确、高效的重载液压机械臂系统逆动力学模型.利用牛顿-欧拉递推法、Spong柔性关节模型和Stribeck摩擦模型建立机械臂动力学模型;利用Creo完成机械臂三维模型搭建并导入ADAMS分析计算.与试验结果对比,确定系数均大于0.6,与无摩擦纯刚体模型计算结果相比有明显提升,验证了建模方法的准确性.

The miniature swimmer made of magnetic elastic material can achieve continuous deformation through the external uniform magnetic field to complete the swimming action. Despite visual feedback improving path tracking accuracy, visual closed-loop robot control is prone to recognition errors and tracking failures under complex backgrounds. Aiming at the above problems, firstly, the obstacle information in complex environments is obtained, and the improved RRT* algorithm（IIC-RRT*） is proposed for path planning. In addition, YOLOv5 based on circular smooth label（CSL-YOLOv5） recognition and tracking algorithm is used to update the central position and rotation angle of miniature swimmers in real time under complex conditions. Using this method, the miniature swimmer is controlled via dual closed loop servos with position and angle against a complex obstacle background. According to the experimental results, the proposed path planning algorithm improves the efficiency and smoothness of the path generation process, while the recognition and tracking algorithm improves the recognition stability and accuracy of the miniature swimmer. This work may provide an innovative idea for the precision control of the magnetic miniature swimmer under complex backgrounds.

拖曳水池试验是研究船舶性能的基本方法,为满足船舶试验的需要,设计了一种适用于试验领域拖曳水池的拖车结构,简述了拖曳水池拖车的基本组成,建立了基于参数设计的完整拖车三维实体模型,并对拖车抗倾覆稳定性进行校核计算,采用有限元软件建立简化后3种强度和刚度提升方案的拖车模型,采用有限元分析方法分别对其进行强度与刚度分析计算,通过对比不同方案的计算结果,得到相对最优方案的模型参数.经分析计算可知,优化后的模型强度和刚度均能够满足实际需求.

新型系泊装置是一种以机械臂吸盘代替传统缆绳的系泊方式,更加高效便捷,同时也更具安全性.通过对船只码头系泊的状况进行水动力分析,得到船体质心的运动响应及载荷数据,进而实现水动力条件下的系泊机械装置动力学仿真分析,得到系泊工况下机械臂各关节的驱动力矩曲线,为之后机械臂控制系统的研究提供数据支撑.

为了解决变量泵负载敏感系统稳定性偏低问题,解决姿态调整机构效率低、精度低、可操作性差等问题,设计基于负载敏感原理的特种车液压动力系统.介绍系统工作原理,并对液压系统主要元件进行设计计算与选型.选择比例变量泵做动力源,提高了液压动力系统的传动效率、稳定性、调节精度、可操作性等综合性能.

无缆磁控微型机器人由于其尺寸小、质量轻，在生物医疗、微操作领域等都有着广阔的应用前景。设计一种十字叉形磁控微型软体四足机器人，采用水母模式、螺旋桨模式实现机器人的水下运动。首先分别对磁控四足机器人的水母模式和螺旋桨模式进行动力学分析，然后通过静态特性和动态特性分别对两种模式进行仿真与试验的对比分析，研究磁场强度与弯曲特性的关系、水母模式拍打幅度、频率与速度的关系，螺旋桨模式步长与圆锥素线角、频率与速度的关系等。在此基础上，通过设计控制信号分别在两种模式下实现在水下的垂直向上、悬停、水平移动等多个动作，并实现简单路径跟踪试验，磁控微型四足水下机器人的仿真与试验为微型机器人的应用提供了新思路。

对重载机械臂的三种典型工况进行有限元静力学分析,得到最危险工况.在满足强度和刚度的条件下,以伸缩臂轻量化为目标,对其截面尺寸进行优化.根据Box-Behnken试验方法对截面尺寸进行三因素三水平设计参数组合,采用含有交叉项的二次多项式构建响应面函数,利用最小二乘法进行回归,分析各参数与响应面的关系,得到截面最优参数组合.采用ADAMS软件对重载机械臂刚柔耦合运动轨迹进行仿真.结果表明:伸缩臂优化后质量减轻了65.1kg,重载机械臂末端轨迹总误差减小了6.1mm,验证了响应面优化结果的可行性.

重载机械臂电液伺服系统具有高度非线性和模型不确定性的特点,且受摩擦影响大,为使其具有良好的跟踪性能和抗干扰能力,建立了包含Stribeck摩擦模型的重载机械臂俯仰缸电液伺服系统数学模型,提出了一种基于扩张状态观测器的电液伺服系统自适应鲁棒控制策略.将扩张状态观测器和采用反演法设计的自适应鲁棒控制器相融合,对系统中存在的不匹配干扰和未知参数进行估计,并利用Lyapunov定理对系统稳定性进行分析证明.通过MATLAB仿真验证,表明该控制策略能够准确跟踪系统指令并估计出非匹配干扰.与传统自适应鲁棒控制策略相比,该控制器可以有效抑制变负载和未知扰动的影响,有效提高了电液伺服系统的控制精度,能够满足重载机械臂的设计要求.

为解决车载液压行车发电瞬态指标控制问题,设计基于能量调节的泵阀并联液压马达调速系统.研究能量调节技术在车载液压发电系统中的应用,对液压系统主要元件进行选型.参考标准GB/T 2820.5—2009,进行液压发电试验,并对比采用能量调节技术前后得到的试验数据.结果表明:应用能量调节技术的车载液压发电装置,其主要交流发电技术指标达到交流发电机组对应相关G2国家标准;采用能量调节技术能够有效降低液压马达转速波动幅度、提高车载液压行车发电瞬态指标.

针对水体运动导致电子传感装置测量结果准确性下降的问题和阈值分割法无法在光照场景下测量波面的问题,文章提出了一种基于U-net卷积神经网络的波浪测量方法.实验过程首先由高清摄像机录制水槽中的波浪运动过程,将视频处理成时间连续的波面图像,其次通过U-net卷积神经网络对波面图像进行图像分割并提取水位线数据信息,最后求出波高和周期.以像素识别结果为基准,将本研究方法的测量结果与波高传感器的测量结果进行误差对比,结果表明U-net卷积神经网络的相对误差最大为2.25％,而传感器误差最大为4.15％,且实验组中U-net卷积神经网络测得平均波高的相对误差均在2.5％以内,平均周期的误差都低于1％.因此,基于U-net卷积神经网络的测量方法可用于实验室的波浪测量.

In order to study the internal flow state of the mixed-flow turbine, this paper takes the modified HLTZ60-LJ-233 model turbine as the research object, combines with the modeling software UG, uses the grid division software ICEM, and then imports the flow field analysis software Fluent, and the turbine full-flow flow field analysis based on the k- ω turbulence model.

基于图像识别的波浪爬高测量方法,根据给定的采样频率完成对采集视频的帧提取,建立时间序列图像集.通过对图像标定、灰度化、阈值计算和波面识别获取波面数据,进行波浪爬高计算,得到波浪爬高的测量结果.与电容式波高传感器测量结果对比表明,利用文中所述算法获得波浪爬高时间曲线与波高传感器实测曲线具有较好的一致性,且对爬高峰值的测量更加准确,并采用图像识别数据对传感器数据进行修正,能较好消除波高传感器斜置状态下测量形成的误差,在兼顾效率的同时提高了爬高测量精度.

基于磁弹性复合材料设计一种毫米级微型游泳机器人,利用三维亥姆霍兹线圈实现空间均匀磁场对机器人的无缆驱动,通过时变旋转磁场来改变机器人的运动姿态,从而实现机器人的游泳动作.介绍微型游泳机器人结构及制作流程,并利用Abaqus建立仿真模型,通过有限元仿真和试验对机器人的游泳特性进行对比分析,研究磁场频率、强度对游泳速度的影响.在此基础上,采用改进型视距导航控制方法提高游泳机器人的抗干扰能力,同时提出一种速度模糊自适应算法和头尾转换控制实现了对复杂规划路径的精确跟踪,试验结果表明,该算法能克服常规匀速视距导航控制易发生跟踪点丢失、无法再次回归跟踪路径等缺点,并能很好地减小路径跟踪误差,为磁控微型游泳机器人的精确控制提供了新思路.