关于物流机器人配送在社区疫情隔离中的应用探讨

机器人系统通常用于人工介入成本过高、危险过大或者效率过低的任务。近年来,随着人工智能、自动驾驶、物联网等众多相关学科的全面发展,机器人在很多行业的应用也越来越广泛、成熟;与民众生活密切相关的物流机器人的配送应用也在逐步发展;互联网下,电商的充分发展为物流机器人的配送应用提供了明确的应用场景。而在抗击新型冠状病毒肺炎疫情时期,区域人群需要实行严格隔离,隔断病毒人传人的感染途径。针对这样的状况,物流机器人可发挥其巨大优势,为民众提供物资配送服务,可有效隔断病毒人际传播,避免人员配送过程中可能带来的病毒扩散风险,针对性地解决疫情隔离阶段物资供给最后一里的难题。     

动态环境下改进人工势场法的仓储机器人自主导航系统研究

为了解决仓储机器人在全动态环境中的自主导航问题,在分析自主导航技术基础上建立了机器人和动态障碍物的数学模型,搭建了以二维激光雷达为主的环境感知平台,提出了一种改进的人工势场法。在传统人工势场法中同时引入相对速度和相对加速度因素得到改进的人工势场模型,实现机器人在全动态环境中的自主移动。设计了无障碍物和多动态障碍物两种移动环境。经仿真验证,应用改进的人工势场法进行路径规划能高效地避开动态障碍物、跟踪动态目标,且运动路径光滑。     

面向NVM存储系统的快速文件访问系统

NVM存储设备系统具备提供高吞吐的潜质,包括接近内存的读写速度、字节寻址特性和支持多路转发等优势。但现有的系统软件栈并没有针对NVM去设计,使得系统软件栈存在许多影响系统访问性能的因素。通过分析发现文件系统的锁机制具有较大的开销,这使得数据的并发访问在多核心环境下成为一个难题。为了缓解这些问题,设计了无锁的文件读写机制以及基于字节的读写接口。通过取消基于文件的锁机制改变了粗粒度的访问控制,利用自主管理请求提高了进程的并发度;在设计能够利用字节寻址的新的文件访问接口时,不仅考虑了NVM存储设备的读写非对称,还考虑了其读写操作的不同特性。这些设计减少了软件栈的开销,有利于发挥NVM特性来提供一个高并发、高吞吐和耐久的存储系统。最后利用开源NVM模拟器PMEM实现了FPMRW原型系统,使用Filebench通用测试工具对FPMRW进行测试与分析,结果显示,FPMRW相对EXT+PMEM和XFS+PMEM能提高3%~40%的系统吞吐率。     

基于人脸识别与光流追踪的移动机器人视觉导航方法

针对移动机器人视觉导航中跟踪目标丢失的问题,提出了基于人脸识别与稀疏光流算法(KLT)结合的移动机器人视觉导航方法(FR-KLT视觉导航方法)。采用OpenCV库中的Haar特征提取人脸识别算法实时检测识别目标人脸,通过Harris角点检测获取目标人体特征点,对目标人体进行精准定位;KLT光流追踪法测算目标移动趋势,并预测目标下一刻大致位置。目标人体位置变动时移动机器人对目标进行实时追踪导航。通过Pioneer-LX机器人在真实环境下试验,验证了该方法准确识别并跟踪目标的实时性和有效性。     

煤矿井下探测搜救机器人地形感知系统及路径规划方法研究

针对常规机器人导航系统采用单一类型地形识别传感器,观察维度单一等问题,对煤矿井下探测搜救机器人地形感知系统进行研究,使用远近感知系统数据融合,提高机器人避障能力。由激光扫描仪采集的二维点云数据建立远距离地形信息,由Kinect相机采集的地形深度信息建立近距离地形信息。基于PCL模型,应用像素遍方法,实现观测信息的采集与云图像的构建。使用2.5维栅格地图构建方法得到近距离环境地形信息。使用Dijkstra算法进行了路径规划研究,建立了融合路径长度和地面危险度等级的目标函数。通过仿真研究验证了本文提出的最优路径减小机器人行走过程的俯仰角、侧倾角的波动幅度。     

一种新的电子罗盘校准算法研究

针对复杂磁环境下磁强计误差补偿算法效果不理想而导致电子罗盘精度较低的问题,本文对磁强计的误差来源进行了详细分析与建模,提出了一种基于模拟退火算法(SA)的空间椭球磁强计校正方法,采用模拟退火算法,对磁强计测量数据进行空间椭球拟合,用估计的参数进行刻度系数与软磁干扰、硬磁干扰与零点偏移的整体误差补偿。最后,利用最小二乘法求解出非正交轴的方向余弦,进行非正交误差和安装误差补偿。最终利用设计的算法对某电子罗盘系统进行了航向角误差对比试验,实验结果表明该方法能准确计算出磁强计的误差参数,实测航向角精度从4.5°提高到0.4°,提高了一个数量级。该方法在电子罗盘的校准中具有了一定的工程参考价值。     

Perception,navigation, and manipulation in the team KAUST approach to the MBZIRC ground robotics challenge

The ground robotics challenge in the Mohammed Bin Zayed International Robotics Challenge required a ground vehicle equipped with a robotic arm to autonomously locate a panel, select a proper size wrench among several options mounted on the panel, and use the wrench to rotate a valve. Autonomy was the critical factor in this challenge, which required the teams to devise algorithms that can operate successfully in a semistructured environment without human supervision. This paper presents the approaches taken by team KAUST to meet this challenge, ranging from in‐house hardware designs to algorithm integration and customization. We separated the whole objective into three interconnected tasks: Navigation, perception, and manipulation. For the navigation task, we developed a basic robotic exploration scheme to find the panel front side where the wrenches were present. For the perception task, we integrated common object detection algorithms with neural networks to identify the proper size wrench precisely. For successful manipulation, we designed and built a custom gripper, which was inspired by the common grasping behavior of a human hand under tight clearance conditions. The modular structure of the proposed approach allowed the team to progress in several subtasks simultaneously. However, the interconnection between the subtasks necessitated a reliable integration framework between these modules for effective implementation. We tuned our algorithms in extensive experimental studies and eventually obtained 10 consecutive successful navigation runs, 96% true wrench detection rate, and high success rate in wrench grasping. Furthermore, successful complete tests proved the reliability and repeatability of our system.     

In‐water visual ship hull inspection using a hover‐capable underwater vehicle with stereo vision

Underwater visual inspection is an important task for checking the structural integrity and biofouling of the ship hull surface to improve the operational safety and efficiency of ships and floating vessels. This paper describes the development of an autonomous in‐water visual inspection system and its application to visual hull inspection of a full‐scale ship. The developed system includes a hardware vehicle platform and software algorithms for autonomous operation of the vehicle. The algorithms for vehicle autonomy consist of the guidance, navigation, and control algorithms for real‐time and onboard operation of the vehicle around the hull surface. The environmental perception of the developed system is mainly based on optical camera images, and various computer vision and optimization algorithms are used for vision‐based navigation and visual mapping. In particular, a stereo camera is installed on the underwater vehicle to estimate instantaneous surface normal vectors, which enables high‐precision navigation and robust visual mapping, not only on flat areas but also over moderately curved hull surface areas. The development process of the vehicle platform and the implemented algorithms are described. The results of the field experiment with a full‐scale ship in a real sea environment are presented to demonstrate the feasibility and practical performance of the developed system.     

Behaviors classification based distance measuring system for pedestrians via a foot‐mounted inertial sensor

In this paper, we develop a foot‐mounted pedestrian navigation system prototype with the emphasis on distance measuring with an inertial measurement unit (IMU) which implies the characteristics of pedestrian gait cycle and thus can be used as a crucial step indicator for distance calculation. Conventional methods for step detection and step length estimation cannot adapt well to the general pedestrian applications since the parameters in these methods may vary for different persons and motions. In this paper, an adaptive time‐ and frequency‐domains joint distance measuring method is proposed by utilizing the means of behaviors classification. Two key issues are studied: step detection and step length determination. For the step detection part, first behavior classification along with state transition strategy is designed to identify typical pedestrian behaviors including standing still, walking, running and irregular swing. Then a four‐stage step detection method is proposed to adaptively determine both step frequency and threshold in a flexible window. Based on the behavior classification results, a two‐segment functional based step length model is established to adapt the walking and running behaviors. Finally, real experiments are carried out to verify our proposed step detection method and step length model. The results show that the proposed method outperforms the existing representative methods and it exhibits the merits of accuracy and adaptability for different persons in real time and significantly improves the accuracy of distance measuring.     

Finite Horizon Maximum Likelihood Estimation for Integrated Navigation with RF Beacon Measurements

We develop a Finite Horizon Maximum Likelihood Estimator (FHMLE) that fuses Inertial Measurement Unit (IMU) and radio frequency (RF) measurements over a sliding window of finite length for three‐dimensional navigation. Available RF data includes pseudo–ranges, angles of transmission (AoT), and Doppler shift measurements. The navigation estimates are obtained by solving a finite‐dimensional nonlinear optimization using a primal‐dual interior point algorithm (PDIP). The benefits of the proposed estimation method are highlighted using simulations results comparing the FHMLE approach with an Unscented Kalman Filter (UKF), in a scenario where an aircraft approaches a carrier, with RF measurements from beacons aboard the carrier, and low‐cost IMU measurements aboard the aircraft. When the Geometric Dilution of Precision is large, we found that the FHMLE is able to achieve smaller estimation errors than the UKF, which tends to carry a bias throughout the trajectory.