您现在的位置: 首页> 研究主题> 观测模型

观测模型

观测模型的相关文献在1985年到2022年内共计94篇,主要集中在自动化技术、计算机技术、无线电电子学、电信技术、测绘学 等领域,其中期刊论文70篇、会议论文6篇、专利文献161893篇;相关期刊56种,包括人口与经济、江苏大学学报(自然科学版)、国防科技大学学报等; 相关会议6种,包括中国全球定位系统技术应用协会2011年年会暨“北斗”产业化和战略新兴产业发展论坛、信息系统协会中国分会第三届学术年会、第三届图像图形技术与应用学术会议等;观测模型的相关文献由280位作者贡献,包括崔祜涛、任诗鹤、佟仲生等。

观测模型—发文量

期刊论文>

论文:70 占比:0.04%

会议论文>

论文:6 占比:0.00%

专利文献>

论文:161893 占比:99.95%

总计:161969篇

观测模型—发文趋势图

观测模型

-研究学者

  • 崔祜涛
  • 任诗鹤
  • 佟仲生
  • 倪向阳
  • 刘世军
  • 刘思晗
  • 刘江山
  • 刘腾
  • 劳世红
  • 叶泽民
  • 期刊论文
  • 会议论文
  • 专利文献

搜索

排序:

年份

    • 任红格; 吴启隆; 史涛
    • 摘要: 针对传统粒子滤波算法在跟踪目标所处环境迁移,目标姿态变化和发生遮挡时容易出现跟踪框漂移现象,提出一种基于灰狼算法优化的粒子滤波跟踪方法(GWOPF)。首先,将全局特征HSV颜色特征和局部特征方向梯度直方图(HOG)特征加权融合建立观测模型;然后,用灰狼算法(GWO)优化粒子滤波算法结构,利用GWO位置更新机制改善粒子空间分布状况,在粒子重采样前进行权值自适应调节,解决原始粒子滤波方法采样时出现的粒子退化问题并优化滤波效果。实验结果表明,改进后的算法在具有挑战的Tiger和Girl视频序列中跟踪成功率分别达到了97.5%和95.0%,单帧处理时间缩短至24.6 ms和18.4 ms,具有较高的跟踪精度和良好的鲁棒性,能够应对跟踪目标发生旋转、部分遮挡等情况以及实时性要求。
    • 蔡小爱; 张海民
    • 摘要: 由于传统方法没有充分考虑加性噪声对图像处理效果的影响,导致去模糊处理后图像易丢失细节信息,峰值信噪比较小以及图像清晰度较差的问题.为此提出基于稀疏约束的离焦图像多尺度去模糊方法.通过建立观测模型,判断图像斑点位置,消除离焦图像中因噪声引起的序列斑点,降低加性噪声的影响程度.进而分离离焦图像中的前景聚焦和离焦背景区域,提取出离焦图像的深度图.最后通过稀疏约束获取模糊核,并运用该模糊核实现对离焦图像的去模糊处理.实验结果表明,所提方法在图像细节保留、清晰度提升和峰值信噪比方面均具有优越性,有效提升了图像的感官效果.
    • 闫立功
    • 摘要: 针对电动机旋转编码器存在的测量精度受环境影响较大、故障率高、价格昂贵的问题,设计基于矿用架线式电动机车的无速度传感器矢量控制系统.在分析异步电动机数学动态模型以及转子磁链矢量控制基本原理的基础上,设计矿用架线式电动机车无速度传感器矢量控制系统,采用电压/电流混合磁链观测模型以及基于反电动势的MRAS转速观测模型对异步电动机的实时转速进行估算.经实验室验证结果表明:所设计的控制系统估算出的异步电动机的实时转速符合精度要求,运行频率高于30 Hz后,误差小于1%.
    • 胡章芳; 曾林全; 罗元; 罗鑫; 赵立明
    • 摘要: 蒙特卡洛定位(MCL)算法存在计算量大、定位精度差的问题,由于二维码具有携带信息的多样性、二维码识别的方便性与易用性的特点,提出一种融入二维码信息的自适应蒙特卡洛定位算法.首先,利用二维码提供的绝对位置信息修正里程计模型的累计误差后进行采样;然后,采用激光传感器提供的观测模型确定粒子的重要性权重;最后,因为重采样部分采用固定样本集会导致大计算量,所以利用Kullback-Leibler距离(KLD)进行重采样,根据粒子在状态空间的分布情况自适应调整下一次迭代所需粒子数,从而减小计算量.基于移动机器人进行的实验结果表明,改进算法与传统蒙特卡洛算法相比定位精度提高了15.09%,时间缩短了15.28%.
    • ZHAO Yan; GAO Guangen; YE Jikun; GAO Yupeng
    • 摘要: 针对拒止环境下,载体导航设备性能受限严重,甚至丧失导航能力的问题,开展抗拒止环境的全源导航子系统模型研究.在明确全源导航内涵的基础上,归纳几种常见的全源导航方式,并引出拒止环境的概念;然后总结多种新兴的全源导航子系统模型;最后对全源导航的关键问题进行探讨.通过研究,为完善载体在复杂的拒止环境中高精度、高可靠性的导航体制提供技术参考.
    • 孙宁; 秦洪懋; 张利; 葛如海
    • 摘要: In view of the inherent defects of traditional collison avoidance systems in respects of perception range, and recognition accuracy etc. due to adopting single sensor for target recognition, a target recognition method based on the information fusion of radar and machine vision is proposed. With the method, after target sequence is obtained, Mahalanobis distance is introduced to conduct observed values matching on the basis of taget level fusion method. Then joint probability data association ( JPDA) algorithm is applied to data fusion, and the observation model and state model of the system are set up to achieve target recognition based on information fusion. The results of verification test show that the method based on radar and camera data can fulfill accurate target recognition and positioning with wider adaptive engineering field.%鉴于传统车辆避撞系统中,因采用单一传感器进行目标识别,在感知范围、识别准确性等方面存在的固有缺陷,本文中提出了一种基于雷达与机器视觉信息融合的目标识别方法.该方法获取目标序列后,在目标级融合方法的基础上,引入马氏距离进行观测值匹配.再应用联合概率数据关联(JPDA)算法进行数据融合,建立系统观测模型与状态模型,从而实现了基于信息融合的目标识别.试验验证结果表明,该方法基于雷达与摄像头数据,可实现目标的准确识别与定位,其工程适应面更广.
    • 刘毅; 谭国俊; 何晓群
    • 摘要: 采用数学模型法对磷酸铁锂电池进行非线性建模,优化了状态模型及观测模型.模型考虑了充放电倍率、温度、老化循环寿命等因素,对电池松弛效应及极化现象影响进行建模补偿,提高了电池建模的准确度,降低了不同条件下因电池模型造成电池荷电状态(SOC)估算的误差影响.在电池模型参数辨识基础上,提出采样自适应Sigma卡尔曼算法构建SOC估算模型,按照非线性模型对状态变量的分布构建Sigma采样序列,采用模型输出残差更新噪声协方差,赋予Sigma采样序列最优估计及噪声的权值,并实现误差量的实时更新,降低计算复杂度.通过持续大电流、间断电流、变电流放电及充电实验条件下的SOC估算对比实验,验证了自适应Sigma卡尔曼算法快速收敛性,数学描述更准确,具备较高的SOC的观测准确度.%The nonlinear model was applied to describe the lithium iron phosphate battery by mathematical model method,and the status model and observation model were optimized.Take into consideration the influences of charge-discharge rate,temperature variation and aging cycle life,the status model was improved.The observation model was also compensated for battery relaxation and polarization effect.Thus,the battery modeling accuracy was enhanced,and state of charge (SOC)estimation error caused by battery model was reduced under different conditions.Then,on the basis of the parameters identification for battery model,an improved adaptive sigma Kalman filter algorithm was proposed to construct the SOC estimation model.According to state variables distribution of the nonlinear model,the sigma sample sequence was built.Each model residual error covariance of output was used to update the covariance of the noise in real time.The optimal estimate sigma sampling sequence and the weights of noise were also reassigned by real-time updates with low computational complexity.The experiments were carried out to estimate the properties by charge,continuous-current discharge,backlash-current discharge and varying current discharge mode.The results verify the rapid convergence and more accurate mathematical description.It is shown that the accuracy of SOC estimation is improved using proposed model and algorithm.
    • 费博雯; 邵良杉; 刘万军
    • 摘要: The traditional tracking method based on sparse representation may tend to be unstable when processing chal-lenging videos and the phenomenon of tracking drifting may occur when the target is occluded during the process of ob-ject tracking. To solve the problem, a novel approach of object tracking based on sparse representation of sub-region matching is proposed. Firstly, the object template is divided into several sub-regions and the observation model is estab-lished by LK image registration algorithm to predict the object motion state of next frame. Then, the region of observation model of prediction is equally divided and the optimal sub-region in the matching process is searched. Finally, by introduc-ing a new template correction mechanism in the process of template update, it overcomes the phenomenon of tracking drifting effectively. The experimental results demonstrate that the proposed tracking algorithm has ideal effect of tracking in the case of object occlusion, motion, illumination and complex background when the multiple object tracking algorithm is tested under different video sequences, and it has well tracking performance compared to traditional tracking method based on sparse representation.%经典稀疏表示目标跟踪算法在处理复杂视频时不免出现跟踪不稳定情况且当目标发生遮挡时易发生漂移现象.针对这一问题,提出一种基于子区域匹配的稀疏表示跟踪算法.首先,将初始目标模板划分为若干子区域,利用LK图像配准算法建立观测模型预测下一帧目标运动状态.然后,对预测的目标模型区域进行同等划分,并在匹配过程中寻找最优子区域.最后,在模板更新过程中引入一种新的模板校正机制,能够有效克服漂移现象.将该算法与多种目标跟踪算法在不同视频序列下进行对比,实验结果表明在目标发生遮挡、运动、光照影响及复杂背景等情况下该算法具有较为理想的跟踪效果,并与经典稀疏表示跟踪算法相比具有较好的跟踪性能.
    • 卢健; 潘峰; 李阳
    • 摘要: 论文利用模糊推理方法提出了一种基于多特征融合的粒子滤波跟踪算法.该算法不但有效继承了传统的固定权值融合方法,并且依据模糊推理对跟踪期间信息的可靠性来输出权值的大小.根据目标模型的形状信息和颜色信息特征的观测似然函数获取各自在跟踪过程中的权值;依据模糊推理,对跟踪期间某一个变化明显而丢失目标的特征信息改变其权值,同时相应的改变另一个特征信息的权值继续无误差的来跟踪目标.与现有的经典的算法相比,提出的算法有更好的跟踪性能及较小的定位误差.实验结果表明了论文所提出算法的有效性.%This paper proposes a modified particle filter algorithm based on multi-feature fusion by using the fuzzy reasoning method.The method effectively follows fusion methods with the fixed weight, at the same time, the fuzzy inference based upon the reliability of tracking information to decide the weight.We seek the each tracking weight ratio according to the observation likelihood function of target model's sharp and color feature.On the basis of fuzzy reasoning, when the feature information changed during the course of tracking, we can change its weight ratio, at the same time change another feature information weight ratio to continue tracking target without error.Compared with the existing classic algorithms, the proposed algorithm has a better tracking performance property and smaller positioning error.The experimental result shows the better effectiveness than the presented algorithm.
  • 查看更多

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号