Preliminary Network Centric Therapy for Machine Learning Classification of Deep Brain Stimulation Status for the Treatment of Parkinson’s Disease with a Conformal Wearable and Wireless Inertial Sensor

The concept of Network Centric Therapy represents an amalgamation of wearable and wireless inertial sensor systems and machine learning with access to a Cloud computing environment. The advent of Network Centric Therapy is highly relevant to the treatment of Parkinson’s disease through deep brain stimulation. Originally wearable and wireless systems for quantifying Parkinson’s disease involved the use a smartphone to quantify hand tremor. Although originally novel, the smartphone has notable issues as a wearable application for quantifying movement disorder tremor. The smartphone has evolved in a pathway that has made the smartphone progressively more cumbersome to mount about the dorsum of the hand. Furthermore, the smartphone utilizes an inertial sensor package that is not certified for medical analysis, and the trial data access a provisional Cloud computing environment through an email account. These concerns are resolved with the recent development of a conformal wearable and wireless inertial sensor system. This conformal wearable and wireless system mounts to the hand with the profile of a bandage by adhesive and accesses a secure Cloud computing environment through a segmented wireless connectivity strategy involving a smartphone and tablet. Additionally, the conformal wearable and wireless system is certified by the FDA of the United States of America for ascertaining medical grade inertial sensor data. These characteristics make the conformal wearable and wireless system uniquely suited for the quantification of Parkinson’s disease treatment through deep brain stimulation. Preliminary evaluation of the conformal wearable and wireless system is demonstrated through the differentiation of deep brain stimulation set to “On” and “Off” status. Based on the robustness of the acceleration signal, this signal was selected to quantify hand tremor for the prescribed deep brain stimulation settings. Machine learning classification using the Waikato Environment for Knowledge Analysis (WEKA) was applied using the multilayer perceptron neural network. The multilayer perceptron neural network achieved considerable classification accuracy for distinguishing between the deep brain stimulation system set to “On” and “Off” status through the quantified acceleration signal data obtained by this recently developed conformal wearable and wireless system. The research achievement establishes a progressive pathway to the future objective of achieving deep brain stimulation capabilities that promote closed-loop acquisition of configuration parameters that are uniquely optimized to the individual through extrinsic means of a highly conformal wearable and wireless inertial sensor system and machine learning with access to Cloud computing resources.

Error Model of Rotary Ring Laser Gyro Inertial Navigation System

To improve the precision of inertial navigation system（INS） during long time operation,the rotation modulated technique（RMT） was employed to modulate the errorr of the inertial sensors into periodically varied signals,and,as a result,to suppress the divergence of INS errors.The principle of the RMT was introduced and the error propagating functions were derived from the rotary navigation equation.Effects of the measurement error for the rotation angle of the platform on the system precision were analyzed.The simulation and experimental results show that the precision of INS was ① dramatically improved with the use of the RMT,and ② hardly reduced when the measurement error for the rotation angle was in arc-second level.The study results offer a theoretical basis for engineering design of rotary INS.

Distinction of an Assortment of Deep Brain Stimulation Parameter Configurations for Treating Parkinson’s Disease Using Machine Learning with Quantification of Tremor Response through a Conformal Wearable and Wireless Inertial Sensor

Deep brain stimulation offers an advanced means of treating Parkinson’s disease in a patient specific context. However, a considerable challenge is the process of ascertaining an optimal parameter configuration. Imperative for the deep brain stimulation parameter optimization process is the quantification of response feedback. As a significant improvement to traditional ordinal scale techniques is the advent of wearable and wireless systems. Recently conformal wearable and wireless systems with a profile on the order of a bandage have been developed. Previous research endeavors have successfully differentiated between deep brain stimulation “On” and “Off” status through quantification using wearable and wireless inertial sensor systems. However, the opportunity exists to further evolve to an objectively quantified response to an assortment of parameter configurations, such as the variation of amplitude, for the deep brain stimulation system. Multiple deep brain stimulation amplitude settings are considered inclusive of “Off” status as a baseline, 1.0 mA, 2.5 mA, and 4.0 mA. The quantified response of this assortment of amplitude settings is acquired through a conformal wearable and wireless inertial sensor system and consolidated using Python software automation to a feature set amenable for machine learning. Five machine learning algorithms are evaluated: J48 decision tree, K-nearest neighbors, support vector machine, logistic regression, and random forest. The performance of these machine learning algorithms is established based on the classification accuracy to distinguish between the deep brain stimulation amplitude settings and the time to develop the machine learning model. The support vector machine achieves the greatest classification accuracy, which is the primary performance parameter, and K-nearest neighbors achieves considerable classification accuracy with minimal time to develop the machine learning model.

A Quaternion Scaled Unscented Kalman Estimator for Inertial Navigation States Determination Using INS/GPS/Magnetometer Fusion

This Inertial Navigation System (INS), Global Positioning System (GPS) and fluxgate magnetometer technologies have been widely used in a variety of positioning and navigation applications. In this paper, a low cost solid state INS/GPS/Magnetometer integrated navigation system has been developed that incorporates measurements from an Inertial Navigation System (INS), Global Positioning System (GPS) and fluxgate magnetometer (Mag.) to provide a reliable complete navigation solution at a high output rate. The body attitude estimates, especially the heading angle, are fundamental challenges in a navigation system. Therefore targeting accurate attitude estimation is considered a significant contribution to the overall navigation error. A better estimation of the body attitude estimates leads to more accurate position and velocity estimation. For that end, the aim of this research is to exploit the magnetometer and accelerometer data in the attitude estimation technique. In this paper, a Scaled Unscented Kalman Filter (SUKF) based on the quaternion concept is designed for the INS/GPS/Mag integrated navigation system under large attitude error conditions. Simulation and experimental results indicate a satisfactory performance of the newly developed model.

基于图优化的GNSS/双目视觉/惯性SLAM系统开发及应用

为提高机器人室外长航时定位精度,提出一种基于图优化的全球导航卫星系统(GNSS)/双目视觉/惯性同时定位与建图(SLAM)系统开发及应用。将空间中的线特征作为几何约束的补充,集成至前端的特征提取及后端的位姿优化线程,提升位姿解算精度。同时,以因子图构建联合优化的图结构,并推导出全局观测误差模型。近200 m的BullDog-CX机器人巡检结果表明,所提算法相比于VINSFusion和PL-VINS分别取得约12.6%及3.4%的定位精度提升,为室外机器人长航时导航提供了一种可行方案。

针对视觉/惯导系统的异常数据检测算法研究

视觉/惯导航系统(VINS)中传感器异常数据的故障检测(FD)方法对提高系统的定位性能和可靠性至关重要。然而,基于视觉/惯导航系统的异常数据检测与排除方法的研究相对较少。提出一种基于传感器测量残差的异常数据检测算法(VINS-ORFD),通过对相机和IMU异常数据进行主动检测并过滤,提升系统可靠性。基于TUM数据集的测试结果表明,该算法不仅可以实现视觉和IMU传感器异常数据快速检测,还能提升至少22.66%的定位精度(RMSE)。

基于平面特征的LIO增强GNSS RTK定位性能研究

针对应用单一传感器的智能导航定位技术在城市复杂环境下难以满足准确、无缝和稳定性能要求的问题,提出一种基于平面特征的雷达惯性里程计(LIO)增强全球卫星导航系统(GNSS)实时动态定位技术(RTK)定位性能的方法:通过滑动窗口构建多帧激光雷达(LiDAR)平面特征的关联;并将RTK与LIO在位置域层面直接融合;最后在开阔环境和城市环境中进行实验。结果表明,RTK/惯性导航系统(INS)/LiDAR组合系统能够在GNSS挑战环境中达到分米级精度;同时,在LiDAR平面特征的约束下,该方法速度和姿态的估计性能也可得到改善。

基于轨道不平顺匹配的多源融合列车定位方法

列车定位是保障铁路安全高效运营的基础,面向先进列车运行控制系统高精度、连续、可靠的位置服务需求,研究高精度、高可靠、低成本的匹配定位源提升列车定位系统的稳健性和准确度具有重要的现实意义。本文提出一种无布设轨道特征信号-轨道不平顺,利用惯性导航系统感知和测量轨道不平顺信号实现匹配定位,结合列车运动约束,实现了无布设情况下的全自主、全天候、全地域的连续可靠的组合定位。为了验证该方案的可行性,本文在真实铁路线路上进行了列车车载试验。试验结果表明:一个典型的战术级IMU可以实现0.6 m(98.46%置信度)的列车定位精度。该方法为多传感器融合的列车定位系统提供了一种新的定位信号源,可以用于提高车载列车定位系统的准确性和稳健性。

基于大数据聚类的飞行探测器复合跟踪控制系统设计

飞行探测器会受到气象条件、磁场等环境干扰,影响探测器的运动轨迹,进而影响跟踪精度;为实现对飞行探测器运动轨迹的实时跟踪与控制,设计基于大数据聚类的飞行探测器复合跟踪控制系统;在复合框架体系中,设置主控芯片、伺服驱动设备两类应用结构,并联合信标机元件,调节微惯性传感器器件的实时连接状态,完成飞行探测器复合跟踪控制系统的硬件设计;完善大数据聚类算法,根据跟踪路径节点标定结果,定义具体的复合跟踪坐标系,完成对飞行探测器复合跟踪的实时控制建模,结合相关硬件,实现基于大数据聚类的飞行探测器复合跟踪控制系统的设计;实验结果表明,所设计系统在跟踪1号、2号目标的跟踪准确率平均值分别为96.5%、97.0%,对目标跟踪时延的平均值为0.9 ms,可以保证基站主机对飞行探测器运动轨迹的实时控制能力。

Highly reliable and selective ethanol sensor based on α-Fe2O3 nanorhombs working in realistic environments

A highly reliable and selective ethanol gas sensor working in realistic environments based on alpha-Fe2O3(α-Fe2O3)nanorhombs is developed. The sensor is fabricated by integrating α-Fe2O3 nanorhombs onto a low power microheater based on micro-electro-mechanical systems(MEMS) technology. The α-Fe2O3 nanorhombs, prepared via a solvothermal method, is characterized by transmission electron microscopy(TEM), Raman spectroscopy, x-ray diffraction(XRD), and x-ray photoelectron spectroscopy(XPS). The sensing performances of the α-Fe2O3 sensor to various toxic gases are investigated. The optimum sensing temperature is found to be about 280℃. The sensor shows excellent selectivity to ethanol.For various ethanol concentrations(1 ppm-20 ppm), the response and recovery times are around 3 s and 15 s at the working temperature of 280℃, respectively. Specifically, the α-Fe2O3 sensor exhibits a response shift less than 6% to ethanol at280℃ when the relative humidity(RH) increases from 30% to 70%. The good tolerance to humidity variation makes the sensor suitable for reliable applications in Internet of Things(IoT) in realistic environments. In addition, the sensor shows great long-term repeatability and stability towards ethanol. A possible gas sensing mechanism is proposed.

A Novel Voronoi Based Particle Filter for Multi-Sensor Data Fusion

Seamless and reliable navigation for civilian/military application is possible by fusing prominent Global Positioning System (GPS) with Inertial Navigation System (INS). This integrated GPS/INS unit exhibits a continuous navigation solution with increased accuracy and reduced uncertainty or ambiguity. In this paper, we propose a novel approach of dynamically creating a Voronoi based Particle Filter (VPF) for integrating INS and GPS data. This filter is based on redistribution of the proposal distribution such that the redistributed particles lie in high likelihood region;thereby increasing the filter accuracy. The usual limitations like degeneracy, sample impoverishment that are seen in conventional particle filter are overcome using our VPF with minimum feasible particles. The small particle size in our methodology reduces the computational load of the filter and makes real-time implementation feasible. Our field test results clearly indicate that the proposed VPF algorithm effectively compensated and reduced positional inaccuracies when GPS data is available. We also present the preliminary results for cases with short GPS outages that occur for low-cost inertial sensors.

面向智能驾驶的高精度多源融合定位综述

随着信息时代各行各业效率的提升,传统的人工驾驶交通系统已逐渐无法满足人们对高效率、低风险交通服务的需求,而智能驾驶技术的出现为这一领域带来了机遇。如今,以自动驾驶为代表的智能驾驶已经成为一种实用的深度交叉技术,其核心模块包括高精度定位、场景感知、决策规划与控制等。定位模块作为智能驾驶系统中最基本、最核心的功能模块,需具备高精度、高可用、低时延的性能特点。当前,结合高精度卫星导航、惯性导航以及环境感知的多源融合技术已成为实现泛在智能驾驶所公认的核心手段,通过充分利用车载传感器的量测信息可以实现精确、可靠的定位服务。从导航定位中常用的传感器技术出发,对当前智能驾驶领域涉及的高精度定位技术进行了全面的回顾,给出了主流的基于滤波和因子图优化的多源融合框架,并对代表性算法进行了整理。最后,总结了现阶段智能驾驶中高精度定位技术的发展现状,并对未来的发展趋势进行了展望。

基于MEMS惯导系统的小口径油气管道中心线测绘技术

针对小口径管道内检测场景,提出了基于MEMS惯导系统的小口径油气管道中心线测绘技术,自研了基于捷联惯导平台的速度、位姿与位置解算算法。根据捷联惯导、轮式里程计的误差方程,使用卡尔曼滤波器对IMU输出信息进行补偿矫正,最后融合地表磁标记信息获得高精度的管道中心线位置信息。利用某内检测项目的实际检测数据进行算法效果验证,并对比了纯惯导方案与多传感器融合方案的效果。结果表明,提出的导航算法可以将检测精度控制在每公里1 m以内,改善了纯惯导测量的检测效果,满足工程应用对检测精度的要求。

基于MEMS-MARG传感器的行人室内定位系统设计

针对低成本行人室内定位系统的实际应用需求,提出了一种基于MEMS-磁、角速度、重力(MARG)传感器的定位方法。基于改进的扩展互补滤波(ECF)算法,结合零速更新(ZUPT)算法与姿态快速收敛(FC)算法,以较低的运算量实现了对行人的定位。设计了相应的软硬件系统,并通过实验对算法性能进行了验证。实验结果表明:基于ECF+FC的行人定位算法性能优于其他基于互补滤波(CF)的方案,能够实现0.5%以内的平面定位误差和1%以内的三维定位误差,可以满足实际需要。

利用惯性传感器数据实时示教下肢外骨骼步态的研究

为实时、灵活和方便地示教下肢外骨骼机器人的步态,提出利用惯性传感器实时示教步态的方法。通过示教者下肢末端的传感器,实时采集脚掌的位姿数据,再根据逆向运动学理论建立下肢各关节角的算法,从而实现每条下肢只需一只传感器即可示教步态的要求。通过安放在小腿和大腿上的其他惯性传感器实测的数据,验证该方法可行性。研究结果表明,该方法时间滞后0.15 s,关节最大误差在2°范围内,能够实现步态的实时示教控制。

视觉惯性导航系统初始化方法综述

视觉惯性导航系统通过初始化,对尺度信息、重力向量、速度、惯性传感器偏差等一系列状态估计所需参数进行快速求解,以提升系统后续导航定位与环境感知的准确性。根据传感信息耦合方式,视觉惯性导航系统初始化方法可以分为三类:联合初始化、非联合初始化和半联合初始化。基于现有研究工作,从基础理论、发展与分类、现有方法、性能评估四个方面展开,对目前主流的初始化方法进行综述,并总结视觉惯性导航系统初始化领域未来的发展趋势,有利于对视觉惯性导航系统初始化方法形成总体性了解并把握其发展方向。

空间引力波探测航天器平台系统前沿研究进展

高精度航天器平台系统是空间引力波探测任务成功实施的重要载体,在空间引力波探测任务中发挥着重要的作用。针对高精度航天器平台系统研究的前沿进展,以惯性传感器参考基准、微推力执行机构、航天器平台无拖曳控制技术及航天器系统编队设计与控制为出发点,结合本专刊内发表的主要研究内容及国内外空间引力波探测航天器平台系统近期相关研究进展,形成涵盖国内外前沿研究热点的概括与综述,总结空间引力波探测航天器平台系统设计的主要研究问题。从航天器系统角度出发,提出涵盖关键载荷、轨道及姿态控制策略及编队构型设计的探测航天器平台系统发展的展望。

卫惯视组合导航技术发展趋势

针对在无人机导航和自动驾驶等领域的大尺度场景中,视惯组合导航系统表现不稳定或因无法闭环而产生误差累积的问题,提出融合卫星导航全局测量优势是实现大尺度室内外无缝导航的重要手段,并从卫星-视觉-惯性(GVI)组合导航数据融合算法角度出发,对GVI组合导航技术研究现状和发展趋势进行总结展望:首先,借鉴同时定位与建图(SLAM)架构,给出GVI组合导航系统的基本理论和框架;接着,从数据融合算法入手,对比分析基于滤波和基于优化方法的方案;然后,探讨不同场景下GVI+组合导航系统在无人系统上的应用潜力;最后,对该领域近年来的发展进行讨论总结并展望发展趋势。

实时水边线船载摄影测量系统研究与开发

针对水边线测绘现有方法存在的高作业风险、高成本等问题,提出一种基于船载摄影的实时水边线测绘新方法。通过数码相机、船用GNSS罗经与IMU集成,研究传感器标定、设备组网、自动化数据采集、数据融合等方法,构建多约束条件的自由网光束法空中三角测量模型,完成实时水边线测绘系统的研发。应用结果表明,100m拍摄距离下,系统绝对定位最大误差为0.258m,中误差为0.128m,精度可满足1:1000及以下比例尺水边线测绘的要求;本系统可切实有效地解决实时水边线测绘现有技术的不足,具有低成本、高精度、高效率、持久续航、一体化等显著优势。

基于惯性导航/无线传感器网络组合的采煤机定位方法研究

采煤机是井下综采工作面的重要设备,采煤机精确定位技术是煤矿生产装备自动化的关键技术之一。为了实现高精度定位满足井下综采自动化作业需求,提出了基于惯性导航/无线传感器网络组合的采煤机定位方法。采用锚节点安置于液压支架上,移动节点与惯导固定安装于采煤机上的配置方案,利用位置已知的锚节点测距信息估计和修正惯导误差,同时实施对安置于推进过程中的液压支架上锚节点(未知节点)位置信息的实时校准,从而达到采煤机高精度定位、无线传感器网络节点动态自动调节的目的。通过试验对所提方法的有效性进行了验证,结果表明,所采用方法对采煤机轨迹具有良好跟踪性能,水平定位误差不超过1.57 cm/h。