Indoor MAV auto-retrieval using fast 6D relocalisation

This paper develops and evaluates methods for performing auto-retrieval of a micro aerial vehicle (MAV) using fast 6D relocalisation from visual features. Auto-retrieval involves a combination of guided operation to direct the vehicle through obstacles using a human pilot and autonomous operation to navigate the vehicle on its return or during re-exploration. This approach is useful in tasks such as industrial inspection and monitoring, and in particular to operate indoors in GPS-denied environments. Our relocalisation methodology contrasts two sources of information: depth data and feature co-visibility, but in a novel manner that validates matches before a RANSAC procedure. The result is the ability of performing 6D relocalisation at an average of 50 Hz on individual maps containing 120 K features. The use of feature co-visibility reduces memory footprint as well as removes the need to employ depth data as used in previous work. This paper concludes with an example of an industrial application involving visual monitoring from a MAV aided by autonomous navigation.     

Development of a search and rescue robot system for the underground building environment

The underground building environment plays an increasingly important role in the construction of modern cities. To deal with possible fires, collapses, and so on, in underground building space, it is a general trend to use rescue robots to replace humans. This paper proposes a dual-robot system solution for search and rescue in an underground building environment. To speed up rescue and search, the two robots focus on different tasks. However, the environmental perception information and location of them are shared. The primary robot is used to quickly explore the environment in a wide range, identify objects, cross difficult obstacles, and so on. The secondary robot is responsible for grabbing, carrying items, clearing obstacles, and so on. In response to the difficulty of rescue caused by unknown scenes, the Lidar, inertial measurement unit and multiview cameras are integrated for large-scale 3D environment mapping. The depth camera detects the objects to be rescued and locate them on the map. A six-degree-of-freedom manipulator with a two-finger gripper is equipped to open doors and clear roadblocks during the rescue. To solve the problem of severe signal attenuation caused by reinforced concrete walls, corners and long-distance transmission, a wireless multinode networking solution is adopted. In the case of a weak wireless signal, the primary robot uses autonomous exploration for environmental perception. Experimental results show the robots' system has high reliability in over-the-horizon maneuvering, teleoperation of the door opening and grasping, object searching, and environmental perception, and can be well applied to underground search and rescue.     

无人系统视觉SLAM技术发展现状简析

视觉SLAM(visual SLAM,vSLAM)已成为无人系统在未知环境中实现全自主导航和环境感知不可或缺的重要组成部分.鉴于此,简要介绍vSLAM的发展历程和典型构成,总结两种前端位姿估计方法(特征法与直接法)和两种后端优化方法(非线性滤波与非线性优化)的典型代表及优缺点.在此基础上,依次按照耦合方式和后端优化方法...     

视觉感知的端到端自动驾驶运动规划综述

视觉感知模块能够利用摄像机等视觉传感器获取丰富的图像和视频信息,进而检测自动驾驶汽车视野中的车辆、行人与交通标识等信息,是自动驾驶最有效、成本最低的感知方式之一。运动规划为自主车辆提供从车辆初始状态到目标状态的一系列运动参数和驾驶动作,而端到端的模型能够直接从感知的数据获取车辆的运动参数,因而受到广泛的关注。为了全面反映视觉感知的端到端自动驾驶运动规划方法的研究进展,本文对国内外公开发表的具有代表性和前沿的论文进行了概述。首先分析端到端方法的应用,以及视觉感知和运动规划在端到端自动驾驶中的作用,然后以自主车辆的学习方式作为分类依据,将视觉感知的端到端自动驾驶运动规划的实现方法分为模仿学习和强化学习两大类,并对各类方法的不同算法进行了归纳和分析;考虑到现阶段端到端模型的研究面临着虚拟到现实的任务,故对基于迁移学习的方法进行了梳理。最后列举与自动驾驶相关的数据集和仿真平台,总结存在的问题和挑战,对未来的发展趋势进行思考和展望。视觉感知的端到端自动驾驶运动规划模型的普适性强且结构简单,这类方法具有广阔的应用前景和研究价值,但是存在不可解释和难以保证绝对安全的问题,未来需要更多的研究改善端到端模型存在的局限性。     

自动驾驶模糊神经网络速度规划方法

为提升自动驾驶的舒适性,降低速度规划算法的复杂度,提出了一种基于模糊神经网络的纵向速度规划方法。将人工驾驶经验总结为模糊规则表,建立了模糊速度规划模型,结合神经网络的自学习功能修正模糊速度规划模型,建立了模糊神经网络速度规划模型。分析了静态障碍物和动态障碍物场景,通过仿真验证了所提速度规划方法的可行性,与传统方法相比,加速度的平滑性能更好。所提速度规划方法具有一定的抗干扰性能,工程实现简单,保证了速度规划的实时性与稳定性。     

Case-based path planning for autonomous underwater vehicles

Case-based reasoning is reasoning based on specific instances of past experience. A new solution is generated by retrieving and adapting an old one which approximately matches the current situation. In this paper, we outline a case-based reasoning scheme for path planning in autonomous underwater vehicle (AUV) missions. An annotated map database is employed to model the navigational environment. Routes which are used in earlier missions are represented as objects in the map. When a new route is to be planned, the path planner retrieves a matching route from the database and modifies it to suit to the current situation. Whenever a matching route is not available, a new route is synthesized based on past cases that describe similar navigational environments. Case-based approach is thus used not only to adapt old routes but also to synthesize new ones. Since the proposed scheme is centered around reuse of old routes, it would be fast especially when long routes need to be generated. Moreover, better reliability of paths can be expected as they are adapted from earlier missions. The scheme is novel and appropriate for AUV mission scenarios. In this paper, we describe the representation of navigation environment including past routes and objects in the navigational space. Further, we discuss the retrieval and repair strategies and the scheme for synthesizing new routes. Sample results of both synthesis and reuse of routes and system performance analysis are also presented. One major advantage of this system is the facility to enrich the map database with new routes as they are generated.This work was supported in part by National Science Foundation Grant No. BCS-9017990.     

Simultaneous exploration and segmentation for search and rescue

We consider the problem of active victim segmentation during a search‐and‐rescue (SAR) exploration mission. The robot is equipped with a multimodal sensor suite consisting of a camera, lidar, and pan‐tilt thermal sensor. The robot enters an unknown scene, builds a 3D model incrementally, and the proposed method simultaneously (a) segments the victims from incomplete multimodal measurements and (b) controls the motion of the thermal camera. Both of these tasks are difficult due to the lack of natural training data and the limited number of real‐world trials. In particular, we overcome the absence of training data for the segmentation task by employing a manually designed generative model, which provides a semisynthetic training data set. The limited number of real‐world trials is tackled by self‐supervised initialization and optimization‐based guiding of the motion control learning. In addition to that, we provide a quantitative evaluation of the proposed method on several real testing scenarios using the real SAR robot. Finally, we also provide a data set which will allow for further development of algorithms on the real data.     

一种移动机器人的禁忌搜索自主导航算法

纯粹的反应式导航算法在复杂未知环境下易陷入局部极小,为此提出一种基于局部子目标和禁忌搜索的自主导航算法．以当前可视区域内障碍物的关键角点为搜索邻域,利用禁忌搜索算法执行优化操作生成当前子目标,进而采用反应式导航算法对其进行跟踪,最终通过子目标的动态切换引导机器人驶达目标位置．算法可有效克服局部极小,显著提高机器人在复杂环境下的自主性．理论分析和仿真实验验证了算法的可行性和有效性．     

An image recognition system aimed at search activities using cyber search and rescue dogs

Disaster response presents major challenges for robotics and computer vision alike. The Cyber‐Enhanced Canine Suit is a suit equipped with a camera, Global Navigation Satellite System (GNSS), and various other sensors, to be worn by search and rescue (SAR) dogs for the purpose of enhancing SAR dog operations. This paper presents an image recognition system for use in disaster scenarios and its integration with the Cyber‐Enhanced Canine Suit platform. The system’s intended use is to spot personal items of missing individuals or other visual clues in video streams from various disaster response platforms. The system facilitates quick learning of targets from limited data and makes providing that data quick and easy. It also provides backtrack recognition functionality, to rapidly find novel targets in the seen footage. We evaluated the recognition system on footage gathered in the field, obtaining promising results. Integrated with the Cyber‐Enhanced Canine Suit, the system can automatically plot detections of search targets onto a map display, to provide operators with a quick overview of what was seen where.     

自主移动机器人室内定位方法研究综述

自主移动机器人的室内定位作为机器人研究领域中最基本的问题已被广泛研究。根据定位技术和传感器的不同,将室内定位方法分为航迹推算定位、地图匹配定位和基于信标定位三类。详细介绍了超声波网络定位系统和基于无线射频识别（RFID）的定位方法。对几种基于概率的定位算法做了分析和对比,并对自主移动机器人室内定位方法的研究方向做了展望。     

Modeling and global trajectory tracking control for an over-actuated MAV

This paper deals with an original micro aerial vehicle (MAV) design, the Omnicopter MAV. It has two central coaxial rotors with fixed-pitch propellers and three perimeter mounted ducted fans with servo motors performing thrust vectoring. Compared with traditional rotary wing MAVs that have inherent underactuation, the Omnicopter possesses some advantages in mobility, for example, lateral translation with zero attitude and hover with nonzero attitude. The trajectory tracking control design, global stability analysis, and control allocation are demonstrated through numerical simulation. The advantage of zero attitude translation is illustrated through experimental results.     

面向自主运行的深空探测航天器体系结构设计 及自主任务规划方法

传统的航天器运行管控方式严重依赖地面指控、人力成本耗费大且对任务响应时间长,难以满足深空探测远距离通信时延突出的任务特点,航天器的智能化水平亟待进一步提高.本文通过对航天器自主运行问题进行系统地分析,提出一种基于多智能体的航天器自主运行的体系结构,面向自主系统结构、运行流程等进行了设计.重点针对小行星探测任务,提出了一种包含平台和载荷任务管理的两阶段航天器自主任务规划算法设计方案,能够实现平载一体的航天器自主任务管理,根据高级任务目标输出完整的指令序列.并通过自主任务规划仿真,对相关算法和模型的正确性和可行性进行了可视化验证和分析,为航天器自主运行技术研究提供了一种有益的思路和方法.     

Heuristic collision-free path planning for an autonomous platform

In this paper, a heuristic and learning, algorithmic scheme for collision-free navigation is presented. This scheme determines an optimum collision-free navigation path of an autonomous platform by using a trial and error process, past navigation knowledge and current information extracted from the generated surrounding environment.     

Radiation search operations using scene understanding with autonomous UAV and UGV

Autonomously searching for hazardous radiation sources requires the ability of the aerial and ground systems to understand the scene they are scouting. In this paper, we present systems, algorithms, and experiments to perform radiation search using unmanned aerial vehicles (UAV) and unmanned ground vehicles (UGV) by employing semantic scene segmentation. The aerial data are used to identify radiological points of interest, generate an orthophoto along with a digital elevation model (DEM) of the scene, and perform semantic segmentation to assign a category (e.g., road, grass) to each pixel in the orthophoto. We perform semantic segmentation by training a model on a dataset of images we collected and annotated, using the model to perform inference on images of the test area unseen to the model, and then refining the results with the DEM to better reason about category predictions at each pixel. We then use all of these outputs to plan a path for a UGV carrying a LiDAR to map the environment and avoid obstacles not present during the flight, and a radiation detector to collect more precise radiation measurements from the ground. Results of the analysis for each scenario tested favorably. We also note that our approach is general and has the potential to work for a variety of different sensing tasks.     

考虑星间测量的航天器自主导航并行滤波器

针对基于星间测量的航天器自主导航问题,本文考虑测量中存在野值的情况,提出了一种轨道根数辅助估计的并行无迹卡尔曼滤波算法.系统由两个并行滤波器组成,通过副滤波器的状态估计识别观测野值,进而在主滤波器中修正导航定位结果.文章选择了星间相对观测两卫星编队的基本构型,研究了算法的阈值参数选择,对不同参数条件下的滤波结果进行了对比.数值仿真说明了该算法在观测量变化率较大时能够有效降低连续野值对自主导航系统的影响,和传统算法相比具有更高的滤波精度和收敛速度.     

Efficient autonomous navigation for planetary rovers with limited resources

Rovers operating on Mars require more and more autonomous features to fulfill their challenging mission requirements. However, the inherent constraints of space systems render the implementation of complex algorithms an expensive and difficult task. In this paper, we propose an architecture for autonomous navigation. Efficient implementations of autonomous features are built on top of the ExoMars path following navigation approach to enhance the safety and traversing capabilities of the rover. These features allow the rover to detect and avoid hazards and perform significantly longer traverses planned by operators on the ground. The efficient navigation approach has been implemented and tested during field test campaigns on a planetary analogue terrain. The experiments evaluated the proposed architecture by autonomously completing several traverses of variable lengths while avoiding hazards. The approach relies only on the optical Localization Cameras stereo bench, a sensor that is found in all current rovers, and potentially allows for computationally inexpensive long‐range autonomous navigation in terrains of medium difficulty.     

不确定海流环境下水下机器人最优时间路径规划

为解决海流预测不精确条件下,现有基于确定性海流路径规划算法鲁棒性差和规划的路径有可能为不可行路径的问题,本文提出一种基于区间优化的水下机器人(AUV)最优时间路径规划算法.该算法采用双层架构,外层用蚁群系统算法(ACS)寻找由起点至终点的候选路径;内层以区间海流为环境模型,计算候选路径航行时间上下限,并分别通过区间序关系和基于可靠性的区间可能度模型将航行时间区间转换为确定性评价函数,并将评价函数值作为候选路径适应度值返回到外层算法.仿真结果表明,相对于确定海流场路径规划方案,提出的方案增强了路径规划器的鲁棒性并解决了结果路径不可行问题.     

自动代客泊车背景下的共享停车供需匹配模型及对应禁忌搜索算法

为了减少自动代客泊车车辆在停车场或指定停车区域内的车辆移位次数和距离,从而降低相关的成本和潜在事故风险,在满足共享停车需求的条件下构建了相应的车辆和泊位匹配优化模型.考虑到无人驾驶车辆停车中可自由移位的特征,将共享停车的需求和供给在时间上加以细分,与决策变量和可行解对应定义了匹配、匹配条和匹配图的概念;通过概念转换将求解匹配图的有效邻居转换为经典指派问题,并利用匈牙利算法加以求解;针对匹配模型的NP-hard特征,设计了对应的禁忌搜索算法.数值分析不仅验证了模型的合理性和求解算法的有效性,也证实模型与方法可处理有人驾驶的共享停车匹配问题.结果表明,利用自动代客泊车可以进一步提升共享泊位利用率,增加可停放的共享车辆数.     

Internet AS幂律建模及其参数估计

为了精确建模Internet自治系统层面上的拓扑结构,提出了基于最小节点度和最大节点度的拓扑幂律模型及其参数估计新算法。针对Internet自治系统层拓扑实际测量数据,利用新算法对拓扑幂律模型中的最小节点度、最大节点度以及标度参数进行计算。实验结果表明,由新算法估计的Internet自治系统层拓扑幂律模型的最小节点度为1,最大节点度随网络规模的增大而增大,标度参数的误差与使用最大然似估计法误差一样均非常小,约为2.25。