基于云和高斯过程的网联车辆协同式道路参数估计
CSTR:
作者:
作者单位:

1.密歇根州立大学 机械工程学院, 东兰辛 48824;2.加利福尼亚大学洛杉矶分校 交通与环境学院, 洛杉矶 90095;3.布法罗大学 机械与航空航天工程学院,布法罗 14260

作者简介:

LI Zhaojian(1988—),男,助理教授,博士生导师,工学博士,主要研究方向为优化控制与估计。 E-mail: lizhaoj1@egr.msu.edu

中图分类号:

U495


Cloud-Based Collaborative Road Profile Estimation Using Gaussian Process
Author:
Affiliation:

1.Department of Mechanical Engineering, Michigan State University, East Lansing 48824, USA;2.Department of Civil and Environmental Engineering, University of California, Los Angeles 90095, USA;3.Department of Mechanical and Aerospace Engineering, University at Buffalo, Buffalo 14260, USA

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [23]
  • |
  • 相似文献 [20]
  • | | |
  • 文章评论
    摘要:

    近年来智能网联汽车发展迅速,云端预先存储的道路参数信息对于提升网联汽车的悬架控制以及检测路面不规则度至关重要。目前关于道路参数估计的工作大多在单个车辆上完成,此类算法对于车辆模型不确定性以及测量误差较敏感。针对该问题,提出了一种新的协同式估计架构,该架构能够充分利用多个同质的网联汽车的测量信息以提高估计精度。首先,在云端利用前方行驶的全部车辆的数据对高斯过程模型进行训练以通过众包方式获取道路参数的估计结果。然后,该结果以未测量的方式发送到后方相邻车辆,后方单个车辆结合自车车载传感器(如加速度计、横摆角速度以及侧倾角速度)和由云端获取的基于众包高斯过程估计结果,使用卡尔曼滤波对该估计结果进一步优化。进而估计结果被发送到云端以更新高斯过程模型。大量的仿真实验结果表明,以该种方式使用云端估计的道路参数作为额外的未测量信息能够提高道路参数的估计精度,验证了该算法的有效性。

    Abstract:

    Road profile information is used to improve vehicle suspension control and detect road irregularities such as potholes. While a great many road profile estimation approaches exist, they have been traditionally performed in a single-vehicle setting, which is inevitably susceptible to vehicle model uncertainty and measurement errors. To overcome these limitations, this paper presents a new collaborative estimation framework that exploits multiple heterogeneous vehicles to iteratively improve the estimation. Specifically, each vehicle combines its onboard measurements (e.g., accelerometers and yaw/roll rate sensors) with a crowdsourced Gaussian process (GP) estimation from the cloud into a Kalman filter (KF) to iteratively refine the estimation. The GP is trained from the crowdsourced local estimations of all prior participating vehicles, which is then sent to the latest participating vehicle as “pseudo-measurements” to enhance the onboard estimation. The resultant local onboard estimation is sent back to the cloud to update the GP. It is shown that using the GP as additional pseudo-measurements can iteratively improve the road profile estimation performance from vehicle by vehicle. Extensive simulations are performed to show the efficacy of the proposed approach.

    参考文献
    [1] LI Z , KOLMANOVSKY I , ATKINS E ,et al .Cloud aided semi-active suspension control[C]// 2014 IEEE Symposium on Computational Intelligence in Vehicles and Transportation Systems (CIVTS). [s.l.]: IEEE, 2014:76-83.
    [2] LI Z , ZHENG M , ZHANG H . Optimization-based unknown input observer for road profile estimation with experimental validation on a suspension station[C]//2019 American Control Conference (ACC). [s.l.]: IEEE, 2019:3829-3834.
    [3] LI Z , KOLMANOVSKY I V , ATKINS E M , et al . Road disturbance estimation and cloud-aided comfort-based route planning[J]. IEEE Transactions on Cybernetics, 2016, 47(11) : 3879.
    [4] WHAIDUZZAMAN M , SOOKHAK M , GAN I A , et al . A survey on vehicular cloud computing[J]. Journal of Network and Computer applications, 2014, 40: 325.
    [5] ZHAO W , ZHANG H , LI Y . Displacement and force coupling control design for automotive active front steering system[J]. Mechanical Systems and Signal Processing, 2018, 106: 76.
    [6] HUBER M F . Recursive gaussian process: On-line regression and learning[J]. Pattern Recognition Letters, 2014,45:85.
    [7] LI Z . Developments in estimation and control for cloud-enabled automotive vehicles[D].Ann Arbor:University of Michigan, 2016.
    [8] RATH J , VELUVOLU K C , DEFOORT M . Estimation of road profile for suspension systems using adaptive super-twisting observer[C]//2014 European Control Conference (ECC). [s.l.] :IEEE,2014 :1675–1680.
    [9] LI Z , KALABI′C U V , KOLMANOVSKY I V , et al .Simultaneous road profile estimation and anomaly detection with an input observer and a jump diffusion process estimator[C]//2016 American Control Conference (ACC). [s.l.]: IEEE,2016 :1693–1698.
    [10] QIN Y , LANGARI R , WANg Z ,et al . Road profile estimation for semi-active suspension using an adaptive kalman filter and an adaptive super-twisting observer[C]// 2017 American Control Conference (ACC). [s.l.]: IEEE,2017: 973–978.
    [11] WANG Z F , DONG M M , QIN Y C , et al . Road profile estimation for suspension system based on the minimum model error criterion combined with a kalman filter[J]. Journal of Vibroengineering, 2017, 19(6): 4550–4572.
    [12] FAURIAT W , MATTRAND C , GAYTON N , et al . Estimation of road profile variability from measured vehicle responses[J].Vehicle System Dynamics, 2016, 54(5): 585.
    [13] ZHENG M , CHEN X , TOMIZUKA M . Extended state observer with phase compensation to estimate and suppress high-frequency disturbances[C]// 2016 American Control Conference (ACC). [s.l.]: IEEE, 2016: 3521–3526.
    [14] ZHENG M , SUN L , TOMIZUKA M . Multi-rate observer based sliding mode control with frequency shaping for vibration suppression beyond nyquist frequency[J]. IFAC-Papers OnLine, 2016, 49(21): 13.
    [15] KO J, FOX D . Gp-bayesfilters: Bayesian filtering using Gaussian process prediction and observation models[J]. Autonomous Robots, 2009, 27(1): 75.
    [16] Highway location marker[EB/OL].[2020-03-20].https://en.wikipedia.org/wiki/Highway location marker.
    [17] ZUO L , ZHANG P S . Energy harvesting, ride comfort, and road handling of regenerative vehicle suspensions[J]. Journal of Vibration and Acoustics, 2013, 135(1) :1.
    [18] ANDREN P . Power spectral density approximations of longitudinal road profiles[J]. International Journal of Vehicle Design, 2006, 40(1/2/3) :2.
    [19] MACKAY D J . Introduction to gaussian processes[J]. NATO ASI Series F Computer and Systems Sciences, 1998, 168 :133.
    [20] QIN Y , LANGARI R , WANG Z , et al . Road profile estimation for semi-active suspension using an adaptive Kalman filter and an adaptive super-twisting observer[C]//2017 American Control Conference.Seattle:[ S.n.],2017: 973–978.
    [21] RASMUSSEN C E . Gaussian processes in machine learning[C]//Summer School on Machine Learning. [s.l.]: Springer, 2003: 63–71.
    [22] KOCIJAN J . Modelling and control of dynamic systems using Gaussian process models[M]. [s.l.]: Springer, 2016.
    [23] SNELSON E , GHAHRAMANI Z . Local and global sparse gaussian process approximations[J]. Artificial Intelligence and Statistics, 2007,2: 524.
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

LI Zhaojian, HAJIDAVALLOO Mohammad R, XIA Xin, ZHENG Minghui.基于云和高斯过程的网联车辆协同式道路参数估计[J].同济大学学报(自然科学版),2022,50(4):489~496

复制
分享
文章指标
  • 点击次数:1368
  • 下载次数: 529
  • HTML阅读次数: 78
  • 引用次数: 0
历史
  • 收稿日期:2021-10-18
  • 在线发布日期: 2022-05-06
文章二维码