一种基于自适应转移概率矩阵的交互多模型跟踪方法
高扬; 毕欣; 杜劲松; 王伟; 田星; 赵越南; 赵乾; 丛日刚; 仝盼盼; 李想
2015-09-02
专利国别中国
专利号CN104880707B
专利类型发明授权
产权排序1
权利人中国科学院沈阳自动化研究所
其他题名Based on self adaptive transition probability of the matrix a model interactive method for tracking
中文摘要本发明涉及一种基于自适应转移概率矩阵的交互多模型跟踪方法,包括以下步骤:首先计算状态估计的交互作用,然后通过卡尔曼滤波或粒子滤波,获得各模型的输出,进而更新模型概率,输出结果,根据模型概率的变化,自适应调节状态转移概率,用于下一时刻跟踪。本发明避免模型的转移概率是先验给定,根据模型概率的变化,自适应调节状态转移概率;本发明可以对目标进行稳定的跟踪,获取目标的准确轨迹,判断目标的运动趋势;本发明提升雷达的跟踪性能。
是否PCT专利
英文摘要This invention claims a method and device based on self-adapting transition probability of the matrix a model interactive tracking method comprises the following steps: Firstly calculating state estimation the interaction of and then the kalman filtering or particle filter so as to obtain the every model and the output of the probability update model output result according to probability module the change of the self-adapting adjusting situation transition probability is used for next time tracking. The invention avoids the mode of transition probabilities is given according to prior probability module the change of the self-adapting adjusting situation transition probability; The invention can be the target to process stable tracking of obtaining the target of accurately judging the track of the object movement trend; This invention of radar so as to improve tracking performance.
公开日期2017-09-26
申请日期2014-11-30
语种中文
专利申请号CN201410715327.7
专利代理沈阳科苑专利商标代理有限公司 21002
内容类型专利
源URL[http://ir.sia.cn/handle/173321/21048]  
专题沈阳自动化研究所_智能检测与装备研究室
推荐引用方式
GB/T 7714
高扬,毕欣,杜劲松,等. 一种基于自适应转移概率矩阵的交互多模型跟踪方法. CN104880707B. 2015-09-02.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace