共查询到20条相似文献，搜索用时 325 毫秒
1.
WU Xue-dong SONG Zhi-huan 《浙江大学学报(A卷英文版)》2007,8(10):1604-1613
Determination of relative three-dimensional (3D) position, orientation, and relative motion between two reference frames is an important problem in robotic guidance, manipulation, and assembly as well as in other fields such as photogrammetry. A solution to pose and motion estimation problem that uses two-dimensional (2D) intensity images from a single camera is de- sirable for real-time applications. The difficulty in performing this measurement is that the process of projecting 3D object features to 2D images is a nonlinear transformation. In this paper, the 3D transformation is modeled as a nonlinear stochastic system with the state estimation providing six degrees-of-freedom motion and position values, using line features in image plane as measuring inputs and dual quaternion to represent both rotation and translation in a unified notation. A filtering method called the Gaussian particle filter (GPF) based on the particle filtering concept is presented for 3D pose and motion estimation of a moving target from monocular image sequences. The method has been implemented with simulated data, and simulation results are provided along with comparisons to the extended Kalman filter (EKF) and the unscented Kalman filter (UKF) to show the relative advantages of the GPF. Simulation results showed that GPF is a superior alternative to EKF and UKF. 相似文献
2.
3.
为了进一步提高移动台的跟踪和定位性能,提出了一种基于联邦滤波结构和简化UKF的移动位置最优估计与融合新方法.该算法以Singer移动台运动模型作为参考系统,以简化UKF滤波器作为子滤波器,对2组独立检测的TDOA和Doppler测量值进行局部估计;然后在主滤波器中,对子滤波器的估计结果按标量加权进行最优融合,得到全局最优或次最优融合估计结果;最后主滤波器利用全局估计结果对子滤波器和参考系统进行反馈和信息重置,以进行下一步估计.仿真试验中,对该算法用于移动台位置估计的效果和性能进行评估,并与基于TDOA和基于Doppler的简化UKF方法做比较.仿真结果表明,该算法能有效降低移动台位置估计的误差和方差,具有良好的均方根误差和均值误差CDF性能. 相似文献
4.
GUO Rong-hua QIN Zheng 《浙江大学学报(A卷英文版)》2007,8(10):1588-1595
In this study, an unscented particle filtering method based on an interacting multiple model (IMM) frame for a Markovian switching system is presented. The method integrates the multiple model (MM) filter with an unscented particle filter (UPF) by an interaction step at the beginning. The framework (interaction/mixing, filtering, and combination) is similar to that in a standard IMM filter, but an UPF is adopted in each model. Therefore, the filtering performance and degeneracy phenomenon of particles are improved. The filtering method addresses nonlinear and/or non-Gaussian tracking problems. Simulation results show that the method has better tracking performance compared with the standard IMM-type filter and IMM particle filter. 相似文献
5.
鲁棒UKF滤波算法在SINS初始对准中的应用 总被引：2，自引：0，他引：2
针对系统存在不确定性扰动时传统UKF滤波算法的滤波精度和鲁棒性均下降的问题,提出了一种基于H∞范数的鲁棒UKF滤波算法.该算法在Krein空间内对简化UKF滤波算法进行改进,增加了一个鲁棒环节.鲁棒环节通过引入给定正常数调整滤波增益从而提高滤波算法的鲁棒性能.在SINS大方位失准角初始对准中对简化UKF滤波算法和鲁棒U... 相似文献
6.
无迹卡尔曼滤波在新型地形无源组合导航系统中的应用 总被引：2，自引：0，他引：2
为了提高自主水下航行器的导航精度,提出了一种新型地形无源组合导航系统.根据水下环境特点和航行器导航高精度和低成本的要求,采用捷联式惯性导航系统、地形辅助导航系统、多普勒速度声纳、电子磁罗经和利用UKF进行信息融合的导航计算机组成新型水下地形无源组合导航系统,并给出了EKF线性滤波方程、UKF非线性滤波方程和导航传感器量测方程.对比仿真实验表明,采用建议的传感器和UKF信息融合方式比采用EKF方式提高了水下航行器导航定位的精度.采用不同的导航传感器和UKF信息融合方法的地形无源组合导航系统可以有效地减小水下航行器导航位置误差,提高组合导航的稳定性和精度. 相似文献
7.
改进粒子滤波在被动目标跟踪中的应用 总被引：3，自引：0，他引：3
As a new method for dealing with any nonlinear or non-Ganssian distributions, based on the Monte Carlo methods and Bayesian filtering, particle filters （PF） are favored by researchers and widely applied in many fields. Based on particle filtering, an improved extended Kalman filter （EKF） proposal distribution is presented. Evaluation of the weights is simplified and other improved techniques including the residual resampling step and Markov Chain Monte Carlo method are introduced for target tracking. Performances of the EKF, basic PF and the improved PF are compared in target tracking examples. The simulation results confirm that the improved particle filter outperforms the others. 相似文献
8.
A method was presented to implement the detecting and tracking of moving targets through omnidirectional vision.The method combined optical flow with particle filter arithmetic.in which optical flow field was used to detect and locate moving targets and particle filter was used to track the detected moving objects.According to the circular image character of omnidirectionaI vision.the calculation equation of optical flow field and the tracking arithmetic of particle filter were improved based on the polar coordinates at the omnidirectional center.The edge of a randomly moving object could be detected by optical flow field and was surrounded by a reference region in the particle filter.A dynamic motion model was established to predict particle state.Histograms were used as the features in the reference region and candidate regions.The mutual information(MI)and Gaussian function were combined to calculate particle weights.Finally,the state of tracked object was computed by the total particle states with weights.Experiment results show that the proposed method could detect and track moving objects with better realtime performance and accuracy. 相似文献
9.
A method was presented to implement the detecting and tracking of moving targets through omnidirectional vision. The method
combined optical flow with particle filter arithmetic, in which optical flow field was used to detect and locate moving targets
and particle filter was used to track the detected moving objects. According to the circular image character of omnidirectional
vision, the calculation equation of optical flow field and the tracking arithmetic of particle filter were improved based
on the polar coordinates at the omnidirectional center. The edge of a randomly moving object could be detected by optical
flow field and was surrounded by a reference region in the particle filter. A dynamic motion model was established to predict
particle state. Histograms were used as the features in the reference region and candidate regions. The mutual information
(MI) and Gaussian function were combined to calculate particle weights. Finally, the state of tracked object was computed
by the total particle states with weights. Experiment results show that the proposed method could detect and track moving
objects with better real-time performance and accuracy.
Supported by Tianjin Higher Education Technology Development Foundation (No.20071308), Tianjin Natural Science Foundation
(06YFJMJC03600) and National Natural Science Foundation of China (No.60773073).
YANG Shuying, born in 1964, female, Dr, Prof. 相似文献
10.
为提高实验教学效果,设计并实现了基于X3DOM的工业机器人虚拟样机,无插件的设计方案使得样机的连接和使用能够通过浏览器进行,降低了对操作环境的要求。其中用X3DOM实现三维模型显示,用HTML5实现控制界面和运动控制算法,二者相结合实现机器人运动学模型到三维模型的映射和虚拟样机的基本功能。此外虚拟样机通过TCP通讯实现了工业机器人的远程控制,能够进行实时的虚实结合实验。最后通过实验验证了机器人虚拟样机的准确性、同步性以及远程控制的实时性。 相似文献
11.
SLAM is one of the most important components in robot navigation. A SLAM algorithm based on image sequences captured by a single digital camera is proposed in this paper. By this algorithm, SIFT feature points are selected and matched between image pairs sequentially. After three images have been captured, the environment＇s 3D map and the camera＇s positions are initialized based on matched feature points and intrinsic parameters of the camera. A robust method is applied to estimate the position and orientation of the camera in the forthcoming images. Finally, a robust adaptive bundle adjustment algorithm is adopted to optimize the environment＇s 3D map and the camera＇s positions simultaneously. Results of quantitative and qualitative experiments show that our algorithm can reconstruct the environment and localize the camera accurately and efficiently. 相似文献
12.
在模式识别领域,基于Unscented的卡尔曼滤波算法（UKF）广受关注,但在求解过程中经常会遇到病态问题,从而影响算法的性能。基于奇异值分解（SVD）的卡尔曼滤波算法（SVDKF）以SVD代替Cholesky分解协方差矩阵产生sigma样本点,可以提高协方差矩阵的数值稳定性。通过对两种算法性能进行仿真比较发现,SVDKF算法优于UKF算法,具有良好的鲁棒性,能有效改善滤波性能,提高算法的精度。 相似文献
13.
Due to the importance of biological safety in drinking water quality and the disadvantages which exist in traditional methods of detecting typical microorganisms such as Cryptosporidium and Giardia, it is necessary to develop an alternative. Particle counts is a qualitative measurement of the amount of dissolved solids in water. The removal rate of particle counts was previously used as an indicator of the effectiveness of a biological activated carbon （BAC） filter in removing Cryptosporidium and Giardia. The particle counts in a BAC filter effluent over one operational period and the effects of BAC filter construction and operational parameters were investigated with a 10 m^3/h pilot plant. The results indicated that the maximum particle count in backwash remnant water was as high as 1296 count/ml and it needed about 1.5 h to reduce from the maximum to less than 50 count/ml. During the standard filtration period, particle counts stay constant at less than 50 count/ml for 5 d except when influenced by sand filter backwash remnant water. The removal rates of particle counts in the BAC filter are related to characteristics of the carbon. For example, a columned carbon and a sand bed removed 33.3% and 8.5% of particles, respectively, while the particle counts in effluent from a cracked BAC filter was higher than that of the influent. There is no significant difference among particle removal rates with different filtration rates. High post-ozone dosage （〉2 mg/L） plays an important role in particle count removal; when the dosage was 3 mg/L, the removal rates by carbon layers and sand beds decreased by 17.5% and increased by 9.5%, respectively, compared with a 2 mg/L dosage. 相似文献
14.
SIFT算法采用的固定阈值在对灰色较多的图像进行匹配时效果较差,通过对SIFT算法进行改进,计算极值点与邻域像素点的灰度差,对灰度差直方图采用迭代算法,得出适应不同对比度的自适应阈值. 相似文献
15.
Implementation of a new PC based controller for a PUMA robot 总被引：1，自引：0，他引：1
FAROOQ M WANG Dao-bo 《浙江大学学报(A卷英文版)》2007,8(12):1962-1970
This paper describes the replacement of a controller for a programmable universal machine for assembly (PUMA) 512 robot with a newly designed PC based (open architecture) controller employing a real-time direct control of six joints. The original structure of the PUMA robot is retained. The hardware of the new controller includes such in-house designed parts as pulse width modulation (PWM) amplifiers, digital and analog controllers, I/O cards, signal conditioner cards, and 16-bit A/D and D/A boards. An Intel Pentium IV industrial computer is used as the central controller. The control software is implemented using VC programming language. The trajectory tracking performance of all six joints is tested at varying velocities. Experimental results show that it is feasible to implement the suggested open architecture platform for PUMA 500 series robots through the software routines running on a PC. By assembling controller from off-the-shell hardware and software components, the benefits of reduced and improved robustness have been realized. 相似文献
16.
无标定情况下的视觉定位是机器人视觉伺服领域的热点和难点。提出一种基于快速自适应滤波的机器人无标定手眼协调方法。首先根据机器人三维基坐标建立运动空间和图像特征空间的微分映射关系,在不预先标定视觉传感器与机器人参数情况下,设计一个直接估计增益矩阵 Kk 的自适应滤波器,在线辨识图像雅可比矩阵。采用最小二乘法获得状态估计的初始值,在此基础上设计视觉控制器和计算运动控制量。在 MAT?LAB 环境下建立机器人手眼协调控制系统仿真模型,将该方法用于观测噪声统计特性未知情况下的定位反馈,并对机器人空间螺旋运动的跟踪情况进行分析。实验结果表明,该自适应滤波器能够完成不规则环境下视觉引导的追踪任务,定位效果优于常规卡尔曼滤波和 Sage-Husa 自适应滤波,具有较强的鲁棒性。 相似文献
17.
This paper presents a disturbance rejection scheme for walking robots under unknown external forces and moments. The disturbance rejection strategy, which combines the inverse dynamics control with the acceleration projection onto the ZMP （zero moment point）-plane, can ensure the overall dynamic stability of the robot during tracking the pre-computed trajectories. Under normal conditions, i.e., the system is dynamically balanced, a primary inverse dynamics control is utilized. In the case that the system becomes unbalanced due to external disturbances, the acceleration projection control （APC） loop, will be activated to keep the dynamic stability of the walking robot through modifying the input torques. The preliminary experimental results on a robot leg demonstrate that the proposed method can actually make the robot keep a stable motion under unknown external perturbations. 相似文献
18.
19.
基于Moravec算子和改进的SIFT算法的图像匹配 总被引：1，自引：0，他引：1
根据sIFT算法具有良好的准确性和鲁棒性,能够克服一定的图像形变及遮挡影响．但其还难以满足图像处理实时性的要求,利用Moravec算子对SIFT算法进行改进,降低SIFT算法特征点获取过程的计算量．通过仿真实验,证明改进的SIFT算法可以提高识别、匹配的速度和准确率,可以满足复杂背景下对图像匹配精度和速度的需求． 相似文献
20.
针对智能停车库中的泊车机器人视觉系统研究需求,提出一种基于双目视觉的泊车机器人障碍物识别系统。通过双目摄像头进行图像采集,利用张正友棋盘标定法进行双目相机标定;采用Bouguet进行立体校正,将高斯滤波与拉普拉斯算子相结合进行图像预处理;采用YOLO卷积神经网络对目标障碍物进行快速识别;利用区域匹配算法进行立体匹配并生成目标障碍物视差图;通过成像点和目标障碍物的立体几何关系计算得到目标障碍物的深度信息。实验结果表明,该系统具有良好的实时性和较高精度,障碍物识别时间平均为0.0901s,在2 600mm具有最佳测距精度,可为泊车机器人自动泊车提供保障。 相似文献