Methodical Approach to the Development of a Radar Sensor Model for the Detection of Urban Traffic Participants Using a Virtual Reality Engine

New approaches for testing of autonomous driving functions are using Virtual Reality (VR) to analyze the behavior of automated vehicles in various scenarios. The real time simulation of the environment sensors is still a challenge. In this paper, the conception, development and validation of an automotive radar raw data sensor model is shown. For the implementation, the Unreal VR engine developed by Epic Games is used. The model consists of a sending antenna, a propagation and a receiving antenna model. The microwave field propagation is simulated by a raytracing approach. It uses the method of shooting and bouncing rays to cover the field. A diffused scattering model is implemented to simulate the influence of rough structures on the reflection of rays. To parameterize the model, simple reflectors are used. The validation is done by a comparison of the measured radar patterns of pedestrians and cyclists with simulated values. The outcome is that the developed model shows valid results, even if it still has deficits in the context of performance. It shows that the bouncing of diffuse scattered field can only be done once. This produces inadequacies in some scenarios. In summary, the paper shows a high potential for real time simulation of radar sensors by using ray tracing in a virtual reality.

BEV-radar:毫米波雷达-相机双向融合的三维目标检测

在自动驾驶场景下的3D目标检测任务中,探索毫米波雷达数据作为RGB图像输入的补充正成为多模态融合的新兴趋势。然而,现有的毫米波雷达-相机融合方法高度依赖于相机的一阶段检测结果,导致整体性能不够理想。本文提供了一种不依赖于相机检测结果的鸟瞰图下双向融合方法(BEV-radar)。对于来自不同域的两个模态的特征,BEV-radar设计了一个双向的基于注意力的融合策略。具体地,以基于BEV的3D目标检测方法为基础,我们的方法使用双向转换器嵌入来自两种模态的信息,并根据后续的卷积块强制执行局部空间关系。嵌入特征后,BEV特征在3D对象预测头中解码。我们在nu Scenes数据集上评估了我们的方法,实现了48.2 m AP和57.6 NDS。结果显示,与仅使用相机的基础模型相比,不仅在精度上有所提升,特别地,速度预测误差项有了相当大的改进。代码开源于https://github.com/Etah0409/BEV-Radar。

Design of Laser Radar Fiber Optic Liquid Level Sensor System

１ＩｎｔｒｏｄｕｃｔｉｏｎＡｓｔｈｅｒａｐｉｄｄｅｖｅｌｏｐｍｅｎｔｏｆｆｉｂｅｒｏｐｔｉｃｓｅｎｓｏｒ［１］，ｌａｓｅｒｒａｄａｒｈａｓｔｒａｄｉｔｉｏｎａｌｙｂｅｎｕｓｅｄｆｏｒｌｏｎｇｒａｎｇｅｄｉｓｔａｎｃｅｍｅａｓｕｒｅｍｅｎｔ．Ｕｓｉｎｇ...

融合毫米波与激光雷达的障碍物检测与跟踪方法

在园区环境中,无人车装载单一毫米波雷达或激光雷达传感器进行障碍物检测跟踪时存在探测范围有限、准确率低及稳定性差等问题。为此,提出一种基于毫米波雷达与激光雷达融合的多障碍物检测跟踪方法。利用改进欧氏聚类算法对道路内激光点云目标进行提取,基于信息筛选策略获得毫米波雷达数据中的有效目标;基于目标检测交并比与可靠性分析,对2种目标进行自适应融合,并利用跟踪门与联合概率数据关联(JPDA)算法完成前后帧数据匹配;应用多运动模型交互与无迹卡尔曼滤波实现障碍物跟踪。实车实验表明:相比单一毫米波雷达与激光雷达障碍物检测跟踪,所提方法有更好的准确性与稳定性。

非接触式生命体征检测装置的设计

为解决临床医学、急救或居家养老等应用场景中无法直接使用传统穿戴式监测仪器实时监测人体生命体征的问题,设计了一种基于多普勒雷达的非接触式生命体征检测装置,该装置通过STM32F429IGT6控制器从多普勒雷达检测的信号中实时提取人体呼吸和心跳信号。其硬件主要由微控制器、多普勒雷达传感器信号放大电路、滤波器、电平抬升电路等部分组成,系统通过IIR带通滤波实现呼吸和心跳信号分离与提取,使用FFT对分离的呼吸和心跳信号进行实时频谱分析,计算并显示呼吸和心跳次数。实验测试结果表明,该装置能有效实时检测人体的呼吸和心跳次数,实现2 m范围内的非接触式人体生命体征检测,心跳次数最大均值误差为9.14%,呼吸次数最大均值误差为4.7%。

基于改进Centerfusion的自动驾驶3D目标检测模型

针对自动驾驶路面上目标漏检和错检的问题,提出一种基于改进Centerfusion的自动驾驶3D目标检测模型。该模型通过将相机信息和雷达特征融合,构成多通道特征数据输入,从而增强目标检测网络的鲁棒性,减少漏检问题;为了能够得到更加准确丰富的3D目标检测信息,引入了改进的注意力机制,用于增强视锥网格中的雷达点云和视觉信息融合;使用改进的损失函数优化边框预测的准确度。在Nuscenes数据集上进行模型验证和对比,实验结果表明,相较于传统的Centerfusion模型,提出的模型平均检测精度均值(mean Average Precision,mAP)提高了1.3%,Nuscenes检测分数(Nuscenes Detection Scores,NDS)提高了1.2%。

基于多传感融合的农田障碍物检测优化研究

为满足农机自主作业过程中障碍物检测的需求,解决视觉检测容易受作业环境条件影响的问题,提出了融合毫米波雷达和相机信息的农田障碍物检测的优化方案。首先,根据雷达散射截面值以及转换矩阵完成雷达点向图像像素坐标系中的映射;然后,利用提出的改进暗通道先验去雾算法及基于显著性的增强算法对映射区域进行处理,以完成局部图像的增强;最后,通过yolov4-tiny网络进行检测,并利用决策级融合策略,完成毫米波雷达同相机检测结果的数据关联,综合输出农田障碍物信息。测试结果表明:相较于仅视觉的检测结果,融入毫米波雷达信息后,在不同的数据集上检测性能均有所提升。在恶劣环境数据集上尤为突出,在晴天有烟尘数据集上,召回率R上升了16.4%,平均精度均值mAP上升了7.95%;在雾天无烟尘数据集上,召回率R上升了17.7%,平均精度均值mAP上升了6.63%。同时,算法对单帧图像的处理时间约148ms,可以满足农机自主作业过程中实时性检测的要求。

一种多传感器骑行安全辅助系统设计

传统自行车缺少灯光指示、环境感知等设备,这为骑行者带来了极大的安全风险和隐患。为此,文中设计了一种多传感器骑行安全辅助系统,该系统以STM32微处理器为主控芯片,以激光雷达、超声波传感器、姿态传感器、霍尔传感器、GPS与通讯模块、LED点阵屏为外围电路,实时感知车辆环境与周边车况,为骑行者提供辅助照明、转向指示、自动鸣笛、碰撞警告、实时定位、危险报警等安全辅助功能。经安装与测试,该系统可提供最大7m的探测距离,低响应延迟,最大25km/h的速度检测,较大提升了骑行者的人身安全。

车载毫米波雷达目标检测综述

目标检测技术作为主动刹车系统的重要组成部分,对测距传感器的检测精度要求极高,毫米波雷达传感器是目标检测的最优选择。综述了毫米波雷达的发展状况、国内外现状、毫米波雷达目标检测算法,介绍了国内外多个重点实验室对毫米波雷达目标检测算法的研究情况,重点放在多传感器融合检测和毫米波雷达多目标检测上。阐述了车载毫米波雷达目标检测的技术难点,对未来毫米波雷达的发展方向提出了建议,以期为后续的研究开发做准备。

基于雷视融合的高速公路交通检测方法

针对单传感器检测的不足和复杂的道路环境,本文设计了视觉传感器与毫米波雷达的车辆目标检测融合算法。通过传感器的参数标定与坐标转换构建了空间融合模型,首先获取了传感器的内参及实际安装的位置信息,然后将毫米波雷达的坐标变换到图像的像素坐标,实现了传感器的时间融合。实验结果表明,与单一传感器的检测结果相比,视觉传感器与毫米波雷达的融合算法能够显著改善视频杂波、雷达的多点分布以及检测目标的分割等问题,目标检测的准确率高于单一传感器,特别是在复杂场景中,融合算法检测的稳定性更高。

基于改进D-S证据理论的毫米波雷达与摄像头融合目标检测和识别

随着汽车保有量的增加和交通环境的日益复杂,汽车行驶安全性和交通事故发生率成为交通安全领域内重点关注的问题,通过智能辅助驾驶技术可以有效提升驾驶安全和降低交通事故发生率。为了提高智能驾驶辅助技术中环境感知的可靠性,提出了一种基于改进D-S证据理论的毫米波雷达与摄像头融合目标检测和识别方法。采用皮尔逊相关性系数改进了D-S证据理论,分析了雷达目标类别与雷达探测信息多参数耦合关系,建立了一种毫米波雷达稀疏点云的目标识别方法。在此基础上,对毫米波雷达信息和视觉检测信息进行加权,提升融合算法在不同工况下目标检测的准确性。实验结果表明,融合算法在不同工况下检测准确率达到93.29%,误检率和漏检率得到降低。相较于其他融合检测算法,本算法在视觉传感器发生漏检时,利用雷达检测算法仍可以对视觉漏检目标进行识别并给出相应的置信度。本工作为毫米波雷达与摄像头的数据融合提供了一种新方法。

Low Power Sensor Design for IoT and Mobile Healthcare Applications

This paper reviews recent advances in radar sensor design for low-power healthcare,indoor real-time positioning and other applications of IoT.Various radar front-end architectures and digital processing methods are proposed to improve the detection performance including detection accuracy,detection range and power consumption.While many of the reported designs were prototypes for concept verification,several integrated radar systems have been demonstrated with reliable measured results with demo systems.A performance comparison of latest radar chip designs has been provided to show their features of different architectures.With great development of IoT,short-range low-power radar sensors for healthcare and indoor positioning applications will attract more and more research interests in the near future.

一种面向多源传感器的飞行器类型识别综合仿真方法

针对现有飞行器类型识别仿真方法主要以单传感器、单目标识别为主,没有考虑真实场景下光电、雷达、电子侦收等多种传感器设备对飞行器参数联合采集的过程,导致仿真场景不丰富、变化不够复杂的问题。构建了面向多源传感器的飞行器综合仿真平台,包括:光电模拟器,使用循环生成对抗网络生成飞行器红外图像;空情模拟器,采用三自由度飞行方法生成飞行器运动轨迹;雷达模拟器,利用椭球方程模拟目标的雷达轨迹和雷达反射截面积;电抗模拟器,通过典型信号源收集和随机噪声引入来模拟复杂电磁环境。在仿真数据集上的实验结果表明,引入多种传感器信息能有效提升目标类型的识别效果,提升模型辅助决策能力。

New structure of Kalman filter for radar networking

Due to the different data rates of the sensors and communication delays in the radar netting, the research of the asynchronous multisensor data fusion problem is more practical than that of the synchronous one. Through discussing the sequential approach, which is the classical asynchronous multisensor data fusion algorithm, a new algorithm based on distributed computation structure is proposed. The new algorithm can meet the requirement of real-time computation of netting fusion system, and is more practical for engineering compared with the classical sequential approach. Simulation results show the validity of the presented algorithm.

Track-to-Track Association Technique in Radar Network in the Presence of Systematic Errors

The presence of systematic measuring errors complicates track-to-track association, spatially separates the tracks that correspond to the same true target, and seriously decline the performances of traditional track-to-track association algorithms. Consequently, the influence of radar systematic errors on tracks from different radars, which is described as some rotation and translation, has been analyzed theoretically in this paper. In addition, a novel approach named alignment-correlation method is developed to estimate and reduce this effect, align and correlate tracks accurately without prior registration using phase correlation technique and statistic binary track correlation algorithm. Monte-Carlo simulation results illustrate that the proposed algorithm has good performance in solving the track-to-track association problem with systematic errors in radar network and could provide effective and reliable associated tracks for the next step of registration.

Digital Receiver for Laser Imaging Radar

With the extension of the application domains for laser imaging radar, it is necessary to find a new technical way to obtain high technical performance and adaptive ability. In this paper, A new concept of digital receiver of laser imaging radar system is presented. This digital receiver is defined as a time varying parameter receiver which possesses large dynamics region and time domain filter. The receiver’s mode, component structure as well as every function of its processing are described. The results and laboratorial data show the feasibility of digital reception. Also, it can exploit the inherent nature of laser imaging radar to obtain high probability of detection.

Joint resource allocation scheme for target tracking in distributed MIMO radar systems

A joint resource allocation scheme concerned with the sensor subset,power and bandwidth for range-only target tracking in multiple-input multiple-output(MIMO)radar systems is proposed.By selecting an optimal subset of sensors with the predetermined size and implementing the power allocation and bandwidth strategies among them,this algorithm can help achieving a better performance within the same resource constraints.Firstly,the Bayesian Cramer-Rao bound(BCRB)is derived from it.Secondly,a criterion for minimizing the BCRB at the target location among all targets tracking in a certain range is derived.Thirdly,the optimization problem involved with three variable vectors is formulated,which can be simplified by deriving the relationship between the optimal power allocation vector and the bandwidth allocation vector.Then,the simplified optimization problem is solved by the cyclic minimization algorithm incorporated with the sequential parametric convex approximation(SPCA)algorithm.Finally,the validity of the proposed method is demonstrated with simulation results.

基于毫米波传感器与激光雷达信号融合的自动驾驶障碍物感知方法

由于自动驾驶无法准确感知障碍物信息,且感知误差较高,为了提高自动驾驶避障能力,提出基于毫米波传感器与激光雷达信号融合的自动驾驶障碍物感知方法。将毫米波传感器内置于自动驾驶装置中,采集自动驾驶装置与障碍物之间的激光雷达回波信号,采用匹配滤波器进行干扰滤波处理,提取传感信号的功率谱密度特征量,采用小波多尺度分解的方法实现信号时频转换,通过Wigner-Vill分布检测和多分辨特征聚类,分析自动驾驶激光雷达毫米波传感信号回波特征,根据波束形成和信号融合结果实现对障碍物感知和自适应定位。测试结果表明,采用该方法进行自动驾驶障碍物感知的准确性较高,感知误差较低,最低为0.01,响应速度较快,最快为0.1 s,收敛迭代步数较小,定位能力较强,提高了自动驾驶的环境适应性。

面向林业资源防护的CGPSO算法UAV航迹优化应用研究

针对传统PSO无人机航迹规划算法在林业资源防护任务中存在收敛速度慢、易陷入局部最优的问题,提出了一种基于CGPSO的无人机航迹优化算法(cauchy gauss particle swarm optimization, CGPSO)。借助雷达传感器对林间环境进行预检,构建了无人机飞行任务环境模型;引入了自适应惯性权重和融合柯西-高斯变异算子调整粒子群算法,平衡全局-局部收敛速度,优化局部极值问题;综合分析了无人机航迹长度代价、障碍物碰撞代价和高程范围代价,建立了航迹规划适应度函数。仿真结果显示,所规划算法适应度标准差达到了0.148 6,用时54.34 s,相比PSO算法,收敛代价值减少了42%,用时提升了25%,与所有算法相比,整体航迹具有较强的鲁棒性,对环境的适应性更优。因此,采用新规划航迹算法在林区进行林业资源防护工作是可行的。

农业轮式机器人三维环境感知技术研究进展

[目的/意义]作为未来农机装备的研究重点,农业轮式机器人正向着智能化与多功能化的方向发展。三维环境感知技术因其获取的信息量丰富、复杂环境下的鲁棒性和适应性好,成为了农业轮式机器人智能化无人作业的基础与关键,其发展水平直接影响到包括农业轮式机器人在内的无人农机的作业质量与效率。[进展]本文首先总结了农业轮式机器人和农业环境感知技术的发展现状,分析了不同类型农业轮式机器人的使用特点和应用现状。其次分析了在农业轮式机器人上实现三维环境感知所主要使用的感知设备及其对应的关键技术,重点阐述了基于激光雷达、视觉传感器和多传感器融合的农业轮式机器人三维环境感知技术的研究进展。[结论/展望]结合农业作业特点与实际需求,指出了农业轮式机器人三维环境感知技术在适用性、环境信息处理和感知效果等方面存在的一些问题,并提出了提升传感器的农业适用性、发展基于深度学习的农业环境感知技术、发展智能化的高速在线多传感器信息融合技术三个方面的建议,以期为农业轮式机器人三维环境感知技术发展提供参考与借鉴。