共查询到18条相似文献,搜索用时 46 毫秒
1.
2.
3.
4.
5.
针对实际的运动目标跟踪问题中存在的各种物理约束,采用基于在线滚动优化原理的滚动时域估计方法,将跟踪滤波问题转换为带约束的有限时域优化问题,并通过引入到达代价函数,有效减少了优化问题求解所需的计算量。最后,对实际的目标跟踪问题进行了滚动时域估计仿真研究。Monte Carlo仿真结果表明,滚动时域估计能有效提高跟踪精度,并且能在采样周期之内完成求解,满足在线估计的需要。 相似文献
6.
针对视觉里程计易受到动态物体及遮挡因素影响导致的位姿估计准确度低、鲁棒性差等问题,提出一种基于运动约束的无监督视觉里程计算法。首先通过最小重投影误差方法处理了大范围场景中的前景遮挡,考虑到现实场景中的运动物体影响,结合光流估计设计了一种运动掩膜处理方法,有效剔除了场景中动态物体像素信息;其次,对于场景中的重复结构和均匀纹理区域,通过数据驱动从轨迹数据中学习车辆的行为模式建立运动学约束,将车辆运动模型建模为多源时间序列回归模型避免陷入局部解;最后,结合运动掩膜与运动学模型约束所设计的无监督深度学习框架,对单目相机运动位姿及场景深度进行同步估计,提高了位姿估计精度及模型适应性。在KITTI道路公开数据集和校园低速无人车平台上的实验结果表明,所设计算法的位姿估计精度及深度估计精度优于目前主流无监督单目视觉里程计方法。 相似文献
7.
基于三条相互垂直直线的单目位姿估计 总被引:2,自引:0,他引:2
基于单目视觉的位姿估计是计算机视觉中的典型问题之一。文中利用目标物体上的三条相互垂直的直线特征和相机像平面上这些特征的对应获得已标定相机相对于目标物体的位姿参数,给出其闭式求解方法,并证明问题解的数量与相机光心和三条直线的相对位置有关。当光心位于两个特殊平面以外时存在唯一解,反之若在该两个平面之间则存在两个解,并且这两个解具有对称性,该性质可作为合理解的判别依据。由于三条相互垂直的直线是长方体的三条边缘,而长方体在现实世界中广泛存在,该结论为应用直线特征进行单目视觉位姿估计及合作目标设计提供理论依据。 相似文献
8.
针对噪声方差不确定的约束系统,讨论了一种鲁棒滚动时域估计(MHE)方法.首先,根据噪声方差不确定模型,找到满足所有不确定性的最小方差上界,在线性矩阵不等式(LMI)框架下求解优化问题,得到近似到达代价的表达形式;然后再融合预测控制的滚动优化原理,把系统的硬约束直接表述在优化问题中,在线优化性能指标,估计出当前时刻系统的状态.仿真时与鲁棒卡尔曼滤波方法进行比较,结果表明了该方法的有效性. 相似文献
9.
考虑整车主动悬架系统的约束状态估计问题,本文提出基于一致性原理的分布式滚动时域估计(DMHE)算法.首先,为了降低状态估计过程中的计算量,将整车主动悬架系统分解为若干降阶子系统.其次,为提高分布式状态估计效果,采用滚动时域估计(MHE)方法处理主动悬架系统的状态和噪声约束.考虑子系统与邻居估计状态的相关性,在采样间隔中执行多次一致性原理实现主动悬架系统状态的信息融合,进一步建立了算法的稳定性充分条件.最后,通过对比仿真实验验证算法的有效性和优越性. 相似文献
10.
11.
Rodrigo López-NegreteSachin C. Patwardhan Lorenz T. Biegler 《Journal of Process Control》2011,21(6):909-919
Moving Horizon Estimation (MHE) is an efficient state estimation method used for nonlinear systems. Since MHE is optimization-based it provides a good framework to handle bounds and constraints when they are required to obtain good state and parameter estimates. Recent research in this area has been directed to develop computationally efficient algorithms for on-line application. However, an open issue in MHE is related to the approximation of the so-called arrival cost and of the parameters associated with it. The arrival cost is very important since it provides a means to incorporate information from the previous measurements to the current state estimate. It is difficult to calculate the true value of the arrival cost; therefore approximation techniques are commonly applied. The conventional method is to use the Extended Kalman Filter (EKF) to approximate the covariance matrix at the beginning of the prediction horizon. This approximation method assumes that the state estimation error is Gaussian. However, when state estimates are bounded or the system is nonlinear, the distribution of the estimation error becomes non-Gaussian. This introduces errors in the arrival cost term which can be mitigated by using longer horizon lengths. This measure, however, significantly increases the size of the nonlinear optimization problem that needs to be solved on-line at each sampling time. Recently, particle filters and related methods have become popular filtering methods that are based on Monte-Carlo simulations. In this way they implement an optimal recursive Bayesian Filter that takes advantage of particle statistics to determine the probability density properties of the states. In the present work, we exploit the features of these sampling-based methods to approximate the arrival cost parameters in the MHE formulation. Also, we show a way to construct an estimate of the log-likelihood of the conditional density of the states using a Particle Filter (PF), which can be used as an approximation of the arrival cost. In both cases, because particles are being propagated through the nonlinear system, the assumption of Gaussianity of the state estimation error can be dropped. Here we developed and tested EKF and eight different types of sample based filters for updating the arrival cost parameters in the weighted 2-norm approach (see Table 1 for the full list). We compare the use of constrained and unconstrained filters, and note that when bounds are required the constrained particle filters give a better approximation of the arrival cost parameters that improve the performance of MHE. Moreover, we also used PF concepts to directly approximate the negative of the log-likelihood of the conditional density using unconstrained and constrained particle filters to update the importance distribution. Also, we show that a benefit of having a better approximation of the arrival cost is that the horizon length required for the MHE can be significantly smaller than when using the conventional MHE approach. This is illustrated by simulation studies done on benchmark problems proposed in the state estimation literature. 相似文献
12.
This paper is concerned with moving horizon estimation for a class of constrained switching nonlinear systems, where the system mode is regarded as an unknown discrete state to be estimated together with the continuous state. In this work, we establish the observability framework of switching nonlinear systems by proposing a series of concepts about observability and analyzing the properties of such concepts. By fully applying the observability properties, we prove the stability of the proposed moving horizon estimators. Simulation results are reported to verify the derived results. 相似文献
13.
目前针对人体姿态估计的深度神经网络都是在特征图的固定位置上进行采样,无法对人体姿态的几何变换进行建模,当人体实例在尺寸、姿势、拍摄角度等方面发生变化后,网络泛化能力较差.因此,文中提出基于可变形卷积的多人人体姿态估计方法.利用可变形卷积对目标几何变换建模能力较强的特性,设计特征提取模块,可在人体关键点几何变化的条件下保证检测的准确性.为了进一步提高网络性能,利用预训练残差网络.模型的预测值与二维高斯模型生成的真值用于计算损失,并迭代训练模型,能在拍摄视角、附着物及人物尺度变化等复杂条件下有效检测人体关键点.实验表明,文中模型可有效提升人体关键点检测的准确性. 相似文献
14.
本文从计算机视觉的角度,对障碍物检测进行了研究。本文将单目视觉系统简化为摄像机投影模型。通过几何关系推导法建立测距模型,获得图像坐标与世界坐标系之间的转换关系,最后实现障碍物距离的测算。经试验证明,本文选用的测距模型的误差在可接受范围之内。 相似文献
15.
《Robotics and Autonomous Systems》2014,62(10):1581-1596
Localizing small Autonomous Underwater Vehicles (AUVs) that have limited payload and perception capability is of importance to promote popularization of underwater applications. Two different methodologies, filter and optimization based methods, can both be used to address the localization problem. But they are seldom rigorously compared and their relative advantages are rarely established. This paper presents a rigorous investigation on the relationship between these two methods. Based on this examination, a novel cooperative localization algorithm for the scenario where AUVs are localized by using range measurements from a single surface mobile beacon is proposed. The main contribution of this paper is threefold. First, major difference and close connection between filter based method and optimization based Maximum a Posteriori method are explicitly clarified by analytically solving optimization problems. Second, a novel localization algorithm combining a filter based extended Kalman filter and an optimization based Moving Horizon Estimation is developed for three-dimensional underwater localization in real-time and long-term applications. The algorithm allows data fusion of multiple sensors, imposes physical constraints on states and noises, bounds computational complexity, and achieves a compromise between better accuracy and lower computational requirement. Third, observability analysis of single beacon based localization algorithm is conducted in the context of nonlinear discrete time systems and a sufficient condition is derived. The observability and improved localization accuracy of the proposed localization algorithm are verified in a customized underwater simulator by extensive numerical simulations. 相似文献
16.
人体姿态估计是计算机视觉、模式识别领域的重要研究问题,用于将视频图像中的人体骨骼姿态进行检测识别,在人机交互领域有重要应用;针对闸机场景下人群拥挤、遮挡严重的人体姿态估计问题,提出了基于姿态矫正的人体姿态估计网络PCNet;该网络设计了一种融合全局和局部信息的Transformer特征编码模块,并将其引入到模型特征提取骨干网络中提升精度表现;提出基于时空注意力机制的级联结构的姿态矫正模块,对预测的关键点位置进行矫正,修正因遮挡、小尺度目标等引起的误差较大的关键点;将提出的人体姿态估计方法在COCO数据集和CrowdPose数据集上进行实验,实验结果表示,模型效果与主流方法相比在精度和鲁棒性上均得到了提升。 相似文献
17.
快速的人脸轮廓检测及姿态估计算法 总被引:1,自引:0,他引:1
提出一种基于人脸特征区域划分的人脸轮廓检测方法和快速人脸姿态估计方法.该方法根据特征点在人脸的分布情况将人脸划分为9个区域.对于每个选定的区域,首先检测出其初始轮廓线,然后用三次多项式对其进行曲线拟合处理,最后把不同区域的轮廓线连接起来得到完整的人脸轮廓.此外,为了快速、准确地估计出人脸的姿态,本文从人脸的对称性出发,提出了进行人脸姿态估计的面积模型和近似平面模型.实验表明,本文所提出的轮廓检测方法对于复杂背景中具有不同姿态的人脸图像可以得到较满意的检测结果.和其它检测方法相比,本文方法具有模型简单、计算速度快等优点. 相似文献
18.
《Displays》2021
Generating large-scale and high-quality 3D scene reconstruction from monocular images is an essential technical foundation in augmented reality and robotics. However, the apparent shortcomings (e.g., scale ambiguity, dense depth estimation in texture-less areas) make applying monocular 3D reconstruction to real-world practice challenging. In this work, we combine the advantage of deep learning and multi-view geometry to propose RGB-Fusion, which effectively solves the inherent limitations of traditional monocular reconstruction. To eliminate the confinements of tracking accuracy imposed by the prediction deficiency of neural networks, we propose integrating the PnP (Perspective-n-Point) algorithm into the tracking module. We employ 3D ICP (Iterative Closest Point) matching and 2D feature matching to construct separate error terms and jointly optimize them, reducing the dependence on the accuracy of depth prediction and improving pose estimation accuracy. The approximate pose predicted by the neural network is employed as the initial optimization value to avoid the trapping of local minimums. We formulate a depth map refinement strategy based on the uncertainty of the depth value, which can naturally lead to a refined depth map. Through our method, low-uncertainty elements can significantly update the current depth value while avoiding high-uncertainty elements from adversely affecting depth estimation accuracy. Numerical qualitative and quantitative evaluation results of tracking, depth prediction, and 3D reconstruction show that RGB-Fusion exceeds most monocular 3D reconstruction systems. 相似文献