Role-based Bayesian decision framework for autonomous unmanned systems

In the process of performing a task,autonomous unmanned systems face the problem of scene changing,which requires the ability of real-time decision-making under dynamically changing scenes.Therefore,taking the unmanned system coordinative region control operation as an example,this paper combines knowledge representation with probabilistic decisionmaking and proposes a role-based Bayesian decision model for autonomous unmanned systems that integrates scene cognition and individual preferences.Firstly,according to utility value decision theory,the role-based utility value decision model is proposed to realize task coordination according to the preference of the role that individual is assigned.Then,multi-entity Bayesian network is introduced for situation assessment,by which scenes and their uncertainty related to the operation are semantically described,so that the unmanned systems can conduct situation awareness in a set of scenes with uncertainty.Finally,the effectiveness of the proposed method is verified in a virtual task scenario.This research has important reference value for realizing scene cognition,improving cooperative decision-making ability under dynamic scenes,and achieving swarm level autonomy of unmanned systems.

Autonomous Unmanned Aerial Vehicles Based Decision Support System for Weed Management

Recently,autonomous systems become a hot research topic among industrialists and academicians due to their applicability in different domains such as healthcare,agriculture,industrial automation,etc.Among the interesting applications of autonomous systems,their applicability in agricultural sector becomes significant.Autonomous unmanned aerial vehicles(UAVs)can be used for suitable site-specific weed management(SSWM)to improve crop productivity.In spite of substantial advancements in UAV based data collection systems,automated weed detection still remains a tedious task owing to the high resemblance of weeds to the crops.The recently developed deep learning(DL)models have exhibited effective performance in several data classification problems.In this aspect,this paper focuses on the design of autonomous UAVs with decision support system for weed management(AUAV-DSSWM)technique.The proposed AUAV-DSSWM technique intends to identify the weeds by the use of UAV images acquired from the target area.Besides,the AUAV-DSSWM technique primarily performs image acquisition and image pre-processing stages.Moreover,the Adam optimizer with You Only Look Once Object Detector-(YOLOv3)model is applied for the detection of weeds.For the effective classification of weeds and crops,the poor and rich optimization(PRO)algorithm with softmax layer is applied.The design of Adam optimizer and PRO algorithm for the parameter tuning process results in enhanced weed detection performance.A wide range of simulations take place on UAV images and the experimental results exhibit the promising performance of the AUAV-DSSWM technique over the other recent techniques with the accy of 99.23%.

Technology Development of Unmanned Underwater Vehicles (UUVs)

In recent years, the weapon systems have been changing drastically because of the advancement of science technology and the change of military concept of combat. There is an unmanned system at the center of all those changes. Especially, in case of maritime environment, as the center stage of combat has changed from ocean to coastal areas, it is difficult for the existing naval forces to effectively operate in shallow waters. Therefore, unmanned underwater vehicles (UUVs) are being required at an increasing pace. In this paper, we analyze the characteristics of already developed UUVs, which are the key unmanned system of the marine battlefield environment in the future. Through the analysis of development cases and the investigation of the essential technologies, the critical design issues of UUVs are elaborated. We also suggest the future directions of the UUV technologies based on the case analysis.

Analysis of Design Directions for Unmanned Surface Vehicles (USVs)

In recent years, because of the development of marine military science technology, there is a growing interest in the unmanned systems throughout the world. Also, the demand of Unmanned Surface Vehicles (USVs) which can be autonomously operated without the operator intervention is increasing dramatically. The growing interests lie in the facts that those USVs can be manufactured at much lower costs, and can be operated without the human fatigue, while can be sent to the hostile or quite dangerous areas that are inherently unhealthy for human operators. The utilization and the deployment of such vessels will continue to grow in the future. In this paper, along with the technological development of unmanned surface vehicles, we investigate and analyze the cases of already developed platforms and identify the trends of the technological advances. Additionally, we suggest the future directions of development.

AI Safety Approach for Minimizing Collisions in Autonomous Navigation

Autonomous agents can explore the environment around them when equipped with advanced hardware and software systems that help intelligent agents minimize collisions.These systems are developed under the term Artificial Intelligence(AI)safety.AI safety is essential to provide reliable service to consumers in various fields such asmilitary,education,healthcare,and automotive.This paper presents the design of an AI safety algorithmfor safe autonomous navigation using Reinforcement Learning(RL).Machine Learning Agents Toolkit(ML-Agents)was used to train the agentwith a proximal policy optimizer algorithmwith an intrinsic curiositymodule(PPO+ICM).This training aims to improve AI safety and minimize or prevent any mistakes that can cause dangerous collisions by the intelligent agent.Four experiments have been executed to validate the results of our research.The designed algorithmwas tested in a virtual environment with four differentmodels.A comparison was presented in four cases to identify the best-performing model for improvingAI safety.The designed algorithmenabled the intelligent agent to perform the required task safely using RL.A goal collision ratio of 64%was achieved,and the collision incidents were minimized from 134 to 52 in the virtual environment within 30min.

Development of an Inexpensive Decentralized Autonomous Aquatic Craft Swarm System for Ocean Exploration

Swarm robotics in maritime engineering is a promising approach characterized by large numbers of relatively small and inexpensive autonomous aquatic crafts (AACs) to monitor marine environments. Compared with a single, large aquatic manned or unmanned surface vehicle, a highly distributed aquatic swarm system with several AACs features advantages in numerous real-world maritime missions, and its natural potential is qualified for new classes of tasks that uniformly feature low cost and high efficiency through time. This article develops an inexpensive AAC based on an embedded-systemcompanion computer and open-source autopilot, providing a verification platform for education and research on swarm algorithm on water surfaces. A topology communication network, including an inner communication network to exchange information among AACs and an external communication network for monitoring the state of the AAC Swarm System (AACSS), was designed based on the topology built into the Xbee units for the AACSS. In the emergence control network, the transmitter and receiver were coupled to distribute or recover the AAC. The swarm motion behaviors in AAC were resolved into the capabilities of go-to-waypoint and path following, which can be accomplished by two uncoupled controllers: speed controller and heading controller. The good performance of velocity and heading controllers in go-to-waypoint was proven in a series of simulations. Path following was achieved by tracking a set of ordered waypoints in the go-to-waypoint. Finally, a sea trial conducted at the China National Deep Sea Center successfully demonstrated the motion capability of the AAC. The sea trial results showed that the AAC is suited to carry out environmental monitoring tasks by efficiently covering the desired path, allowing for redundancy in the data collection process and tolerating the individual AACs’ path-following offset caused by winds and waves.

无人集群博弈对抗系统仿真验证及决策关键技术综述

无人集群博弈对抗是一种新兴的作战样式,在智能化战争扮演着至关重要的作用,其核心是自主生成博弈对抗决策序列,为集群“赋能”。分析了无人集群博弈对抗系统仿真验证的进展;从基于专家系统和博弈论的技术、基于群体智能和优化理论的技术,以及基于神经网络和强化学习的技术三个方面论述了自主决策关键技术,以及课题组在自主决策上开展的相关工作;提出了无人集群博弈对抗的发展方向。

基于空海异构无人平台的水下目标搜索与跟踪

海上异构无人系统可有效提高复杂任务的完成效率。文中采用自主水下航行器(AUV)和无人机(UAV)来完成近海海域内未知水下目标的搜索与跟踪任务。首先,描述了水下目标搜索跟踪任务,将任务过程分为目标搜索和目标跟踪阶段,2个阶段的目标分别是使AUV&UAV总搜索空间最大化以及AUV与水下目标的末端位置误差最小;然后,建立AUV&UAV跨域协同搜索模型,并设定模型中AUV和UAV探测范围和通信距离等约束条件;最后,在跨域协同搜索与路径跟踪规划中,基于传统粒子群算法,加入自适应学习因子调控策略和精英保存策略,生成搜索与跟踪路径。仿真实验表明,采用改进粒子群优化算法的AUV&UAV异构无人系统能够更高效地完成水下目标搜索与跟踪任务。

中大型无人船舶自主控制系统架构设计及实现

[目的]自主控制系统作为无人船舶的核心系统,负责协调全船其他系统,共同完成各项既定使命任务,是无人船舶技术的重要研究方向。[方法]首先,借鉴主导人类智能的“神经系统”的生理学结构和认知机制,归纳出对于无人船舶自主控制系统架构仿生设计的6点启示。然后,基于这些启示,从“中枢神经系统”和“周围神经系统”两个维度,研究自主控制系统的物理架构及其典型实现。最后,借鉴人类智能的信息处理流程,提出8大功能模块构成的自主控制系统的功能架构以及面向服务架构(SOA)的设计思路。[结果]经研究得到了无人船舶自主控制系统物理架构和功能架构,[结论]其可为中大型无人船舶自主控制系统的总体架构、系统设计与产品研制提供路线参考。

地下空间无人系统研究综述

地下空间作为人类活动空间不可分割的部分,是继宇宙空间、海洋资源外可开拓的第三大领域,也是继陆、海、空、天、电、网之后的第七维战略空间。无人系统通过跨域协同合作可涌现出单个主体难以实现的智能水平,在地下空间发挥关键作用。本文针对地下空间无人系统相关需求,围绕地下空间各领域相关研究进展,梳理其关键技术研究现状及未来发展趋势。介绍了地下矿井无人开采系统、地下轨道交通无人系统、地下管廊综合管理系统、地下空间无人作战系统的国内外研究现状;对地下空间通信、地下空间态势感知、自主导航定位及集群协同控制关键技术进行概述,并指出现存的问题;对地下空间无人系统未来发展趋势进行了展望。

支持隐私保护的自主无人系统分布式智能运维框架

自主无人系统的智能运维是提升系统自主性以及安全可靠运行能力的有效手段,面临分布式运维和隐私保护两个方面的挑战.现有工作未充分考虑自主无人系统的计算资源受限特点及对量化隐私保护的需求,无法应对上述挑战.因此,本文提出了一种支持隐私保护的自主无人系统分布式智能运维框架,基于联邦学习、区块链和隐私保护技术,采用离线学习与在线监测、本地模型与全局模型相结合的方式,利用基于联邦学习的运维模型训练机制,基于区块链的运维模型共享机制,以及基于差分隐私的局部模型共享机制,有效解决系统面临的资源受限和隐私保护问题,为系统提供安全可靠的运维环境以及支持多端融合、自主更新的运维能力.实验结果表明,所设计的框架可以支持高效智能运维方法的构建,针对框架所设计的联邦学习和隐私保护方法是有效的.

水下机器人自主操作--机遇、进展与挑战

21世纪以来,随着人类对海洋资源的探索不断加深,水下非结构化环境中日趋精细复杂的操作任务对潜器的作业能力提出了更高要求和挑战。利用机械臂与周围环境实体进行交互是潜器水下作业能力的重要体现,然而,当前水下自主操作技术尚处于起步阶段,各类干预任务过度依赖人员控制,在连续作业时效率低下,故提高水下作业操作的自主程度是未来海洋装备发展的必然趋势。该文详细介绍了目前水下作业操作技术的发展概况与存在问题,通过对比传统遥控操作手段的局限说明水下机器人自主操作在海洋工程等领域的巨大潜力;综述了水下机器人自主操作技术的研究现状,分析了制约该技术发展的若干因素与挑战,并指明水下机器人自主操作向工程化应用所需突破的关键核心技术;最后,对该领域今后的发展方向作出展望。

基于STPA和Bow-Tie模型的地面无人平台系统安全分析方法

随着地面无人平台(Unmanned Ground Vehicles,UGVs)在复杂作业环境中的潜在应用和战略价值日益凸显,确保其自主行为的安全性变得至关重要。提出一种结合系统理论过程分析(System-Theoretic Process Analysis,STPA)和Bow-Tie模型的地面无人平台系统安全分析方法。围绕遥控操作地面无人平台系统安全,通过STPA方法识别UGV系统中的不安全控制行为及其潜在风险,并利用Bow-Tie模型分析从损失致因场景到可能事故后果的事件链,得到风险传播路径和风险扩散路径。最终,基于Bow-Tie分析结果确定主被动安全分级控制措施,并通过自主安全控制器实现了系统安全管理。

无人自主系统能力边界参数自适应判别方法

为有效应对仿真测试面临的维度灾难问题,降低传统全参数空间遍历中所需的仿真次数,需要获取针对性的仿真数据以准确反映实验数据建模特征,以较少的仿真次数获得信息量丰富且代表原始数据特征的样本。提出一种面向无人自主系统能力边界参数自适应判别的数字化仿真测试模型,采用多权重结构的佳点集进行初始构建,结合自适应核函数边界点判别算法,通过高斯过程回归对模型进行迭代优化,自适应地判别无人自主系统的能力边界。实验结果表明:该方法能够降低建模所需数据量,提高自适应参数边界判别的效率,为提升智能无人系统试验的效率提供了高效途径。

Energy-efficient joint UAV secure communication and 3D trajectory optimization assisted by reconfigurable intelligent surfaces in the presence of eavesdroppers

We consider a scenario where an unmanned aerial vehicle(UAV),a typical unmanned aerial system(UAS),transmits confidential data to a moving ground target in the presence of multiple eavesdroppers.Multiple friendly reconfigurable intelligent surfaces(RISs) help to secure the UAV-target communication and improve the energy efficiency of the UAV.We formulate an optimization problem to minimize the energy consumption of the UAV,subject to the mobility constraint of the UAV and that the achievable secrecy rate at the target is over a given threshold.We present an online planning method following the framework of model predictive control(MPC) to jointly optimize the motion of the UAV and the configurations of the RISs.The effectiveness of the proposed method is validated via computer simulations.

架空输电线路无人机巡检缺陷智能识别技术研究

在电力系统的自主巡检中,无人机需要能够在各种环境条件下捕捉高质量的图像,以便准确地检测和识别潜在的问题。为了实现这一目标,采用改进的Faster R-CNN(Faster Region-Convolutional NeuralNetworks)算法,考虑了不同的拍摄背景、光照条件和拍摄距离等因素,这些都是影响无人机拍摄效果的典型环境工况条件。为了提高识别准确性,对无人机拍摄的图像进行了规范化处理,以确保图像质量满足识别算法的要求。这种改进后的算法能够快速准确地检测图像中的杆塔、绝缘子、金具等典型电力设备部件。开发了一种自适应拍摄算法,该算法能够根据风速、光照等环境因素的影响,自动调整无人机的拍摄参数,确保拍摄的图像居中、清晰可见。通过上述方法,实现了对杆塔、导地线、绝缘子、金具等关键部位的高清自适应对焦拍摄,大大提高了巡检图像的质量。这种结合了先进算法和自适应控制的方法,显著提高了无人机在复杂环境下的自主巡检能力,确保了电力系统的安全稳定运行。

国外海洋自主无人系统跨域协同实践研究

随着单域无人系统复杂环境感知、精准推理决策、协同运动控制等智能化技术的发展,海洋自主无人系统趋向跨域协同方向发展。分析了国外跨域协同概念及其发展动因,梳理了国外空中域、水面域、水下域三类域节点的海上无人装备典型协同运用实践,从中提取了协同控制架构、协同导航定位、协同态势感知和协同任务规划等关键技术,相关成果可为国内海洋自主无人系统跨域协同技术发展提供参考和支撑。

无人自主系统跨域协同作战试验初探

随着战争形态由信息化向智能化转变,利用无人自主系统在多维空间实现跨域协同效能聚合成为联合作战的重要发展方向,无人自主系统跨域协同作战试验是适应战争形态发展需要应运而生的新型作战试验组织形式。梳理了无人自主系统跨域协同作战试验鉴定的发展动态,分析了试验特点规律和试验能力需求。基于系统工程方法,从时间维、逻辑维和知识维构建了无人自主系统跨域协同作战试验系统工程的三维结构模型,为系统化、程序化、规范化开展作战试验提供参考借鉴。

在语义层重构自主无人系统:需求、框架与运行机理

针对无人系统实现高级任务自主性的问题,提出在语义层重构自主无人系统的技术路线。分析了自主无人系统关键能力需求,阐释了语义重构的内涵、优势和运行机理。自主无人系统的语义重构技术路线强调:通过在语义层重构自主无人系统,实现无人系统任务、环境和系统自身的语义化描述;通过构建知识本体实现领域知识的统一表示和群体信息共享,利用语义知识引擎实现群体认知达到任务协同和能力互操作的目的;利用知识推理和不确定性知识推理实现认知基础上的任务自主规划。通过在语义层重构自主无人系统,将自主无人系统的规划问题上升到认知层面,实现硬件与软件,规划问题与规划方法,人类干预与自主规划等问题的解耦。

无人自主系统分布式协同控制研究综述

随着信息技术、人工智能和机器人技术的飞速发展,无人自主系统在军事、航空航天、海洋探索、灾难救援以及智能交通等领域展现出巨大的应用潜力。分布式协同控制作为实现多无人自主系统高效、灵活协作的关键技术,已成为研究的热点。本文综述了无人自主系统分布式协同控制的研究进展。首先探讨了在一致性问题、编队控制和分布式优化三个方面的核心理论,然后结合当前多无人自主系统的实际应用,给出了无人机、无人车、无人水面舰艇、无人潜航器和多模态协同控制的最新研究成果介绍,最后探讨了该领域的未来挑战和发展方向。