基于深度强化学习的智能船舶航迹跟踪控制

[目的]智能船舶的航迹跟踪控制问题往往面临着控制环境复杂、控制器稳定性不高以及大量的算法计算等问题。为实现对航迹跟踪的精准控制,提出一种引入深度强化学习技术的航向控制器。[方法]首先,结合视线(LOS)算法制导,以船舶的操纵特性和控制要求为基础,将航迹跟踪问题建模成马尔可夫决策过程,设计其状态空间、动作空间、奖励函数;然后,使用深度确定性策略梯度(DDPG)算法作为控制器的实现,采用离线学习方法对控制器进行训练;最后,将训练完成的控制器与BP-PID控制器进行对比研究,分析控制效果。[结果]仿真结果表明,设计的深度强化学习控制器可以从训练学习过程中快速收敛达到控制要求,训练后的网络与BP-PID控制器相比跟踪迅速,具有偏航误差小、舵角变化频率小等优点。[结论]研究成果可为智能船舶航迹跟踪控制提供参考。     

铁路沿线城市空间物业开发研究——以宝鸡市南客站核心区站前广场工程为例

为了提升铁路沿线城市空间物业开发的设计手法,结合宝鸡市南客站核心区站前广场工程的设计实践,从系统性设计观、工程性质解析、消防设计思路、空间建构方法、无缝接驳、技术要素等方面进行研究分析,展示将铁路与城市系统进行一体化设计、利用自然资源解决消防和投资问题、平层空间中对多种交通体系的有机组织、不同坐标和高程系统背景下分区域构建柱网、各板块交接处的问题解决等技术成果,总结出同类型建筑的设计思路,力求为该领域的建筑设计发展提供有益的借鉴。     

铁路预应力路堤加固技术数值研究

研究目的:铁路预应力路堤在国内外尚属一种新型路基加固法,其受力变形特性暂未得到系统化研究,相关加固设计理论仍处于探索性阶段。因此,有必要通过数值手段了解预应力路堤的工作状态,以掌握其加固性能。鉴于此,借助ABAQUS软件平台构建预应力路堤仿真系统,分析差异化预应力加固参数对路堤变形和承载能力的影响以及预应力加固构件的受力特征。研究结论:(1)路堤本体段坡面较优加固位置为距本体顶面以下0.3倍本体高度处;(2)坡率1∶1的预应力路堤在第1、2排侧压板分别施加50 k Pa、100 k Pa预压荷载时,其变形与承载力均可达传统路堤(坡率1∶1.5)水平,并可通过提升加固标准进一步强化路堤承载性能;(3)当对第1、2、3排侧压板分别施加50 k Pa、100 k Pa、100 k Pa预压荷载时,路堤内部附加围压S11>13.5 k Pa区域大致呈"x"形分布并形成横贯路堤的"预压加固区";(4)侧压板锚固区受力集中且复杂,应注意保障锚固区板体强度;(5)力筋在路堤加载前后的应力变化量与坡面侧向变形特征相关;(6)本研究成果可为铁路预应力路堤的加固设计提供技术指导。     

基于职业能力的高职会计实践教学体系的重构

文章通过分析高职会计专业构建实践教学体系的必要性入手，分析会计职业能力的内涵，从会计课程体系、实训项目体系、实训项目教学设计三个方面探索了现代高职会计实践教学体系的构建，拓展会计实践教学的实训资源，完善构建实践教学体系的机制。     

递进式连锁经营管理专业实践教学改革研究——基于加盟创业项目视角

连锁经营管理专业人才培养目标大多定位在一线服务员或基层管理者，所培养的学生必须具备较强的连锁专业能力和创业能力，而大多数高职院校培养出来的学生在这些方面所表现出来的能力却差强人意，尤其是创业能力方面。本文提出将加盟创业项目贯穿于连锁经营管理实践教学全过程，按照“仿真训练-校内孵化-校内模拟-市场全真”四个阶段，形成一个递进式的实践教学流程，从而促使学生的专业能力和创业能力实现递进式提升。     

合理养生预防半刚性基层的反射裂缝

半刚性基层的反射裂缝的产生和很多因素有关,文中主要研究分析如何通过科学合理的养生来最大限度地防止反射裂缝的产生。并通过试验分析数据,理论分析,得出土工布养生为较好的养生方法,并将之应用于广河高速S27合同段中,取得了良好的效果。     

公路地基真空排水预压法探讨

真空排水预压法与堆载预压法相比具有加载快、无需堆载材料、不会失稳等优势,适用于加固软土特别是含水量较高的超软土地基。真空-堆载联合预压法的应用也越来越多。经过国内外十几年的探索,真空排水预压法日趋成熟,已经成为大面积应用的加固软土地基的行之有效的方法之一。     

Design,analysis and performance evaluation of a third order distributed protocol for platooning in the presence of time-varying delays and switching topologies

This paper is concerned with the problem of designing a decentralized consensus protocol for platooning of non-identical vehicles in the presence of heterogeneous time-varying communication delays. The proposed control protocol makes use of a state feedback and to this aim drivetrain dynamics are modeled as third-order linear systems. Necessary and sufficient conditions for convergence and exponential stability, derived by using an appropriate Krasovskii functional, demonstrate the ability of the platoon in reaching the required regime with an exponentially bounded behavior. The proposed LMI-based approach allows to estimate both delay margin and decay rate. Moreover, convergence is proven under switching communication network topologies by means of a Lyapunov-Razumikhin function, and the assessment of a string stable behavior has been also theoretically investigated. High-fidelity simulations with Plexe show the effectiveness of the theoretical results in different driving conditions and in the presence of external disturbances and communication impairment. Different communication channel models are used in the validation stage to further prove robustness of the proposed methodology with respect hard delay and packets losses.     

Human-like autonomous car-following model with deep reinforcement learning

This study proposes a framework for human-like autonomous car-following planning based on deep reinforcement learning (deep RL). Historical driving data are fed into a simulation environment where an RL agent learns from trial and error interactions based on a reward function that signals how much the agent deviates from the empirical data. Through these interactions, an optimal policy, or car-following model that maps in a human-like way from speed, relative speed between a lead and following vehicle, and inter-vehicle spacing to acceleration of a following vehicle is finally obtained. The model can be continuously updated when more data are fed in. Two thousand car-following periods extracted from the 2015 Shanghai Naturalistic Driving Study were used to train the model and compare its performance with that of traditional and recent data-driven car-following models. As shown by this study’s results, a deep deterministic policy gradient car-following model that uses disparity between simulated and observed speed as the reward function and considers a reaction delay of 1 s, denoted as DDPGvRT, can reproduce human-like car-following behavior with higher accuracy than traditional and recent data-driven car-following models. Specifically, the DDPGvRT model has a spacing validation error of 18% and speed validation error of 5%, which are less than those of other models, including the intelligent driver model, models based on locally weighted regression, and conventional neural network-based models. Moreover, the DDPGvRT demonstrates good capability of generalization to various driving situations and can adapt to different drivers by continuously learning. This study demonstrates that reinforcement learning methodology can offer insight into driver behavior and can contribute to the development of human-like autonomous driving algorithms and traffic-flow models.     

桥梁单板受力的原因及解决措施

随着交通量的迅猛增长,桥梁单板受力情况越来越频繁,通过对单板受力现象产生原因进行分析,提出解决措施,以供同行参考和借鉴。