首页|一类基于模糊推理的具有机动自适应的目标跟踪算法

一类基于模糊推理的具有机动自适应的目标跟踪算法

扫码查看
针对变结构多模型算法在机动 目标跟踪中对目标机动不确定性、量测不确定性自适应能力不足的问题,提出一种基于模糊推理的机动自适应目标跟踪算法。设计一种基于模糊推理的双级机动判别模型,利用模型概率信息和主模型滤波残差加权范数进行主模型可信度和机动判别推理;并将双级机动判别引入基于可能模型集的期望模式扩增方法(EMA-LMS)框架,提出一种模糊推理EMA-LMS算法,实现对模型集自适应的参数和策略的在线调节,从而生成更加接近目标真实运动模式的期望模型,并更好地对模型进行取舍。仿真结果表明,本文算法能够有效增强算法对目标机动和量测不确定的自适应性,提高跟踪精度。
An Adaptive Maneuvering Target Tracking Algorithm Based on Fuzzy Inference
An adaptive maneuvering target tracking algorithm based on fuzzy inference is proposed to deal with the low adaptive capacity of variable structure interacting multi-model algorithms for target maneuver uncertainty and measurement uncertainty in maneuvering target tracking.A two-stage maneuvering discrimination model based on fuzzy inference is designed,which uses the probability of models and residual weighted norm of the main model to infer the reliability of the main model and the possibility of maneuvering discrimination.The two-stage maneuvering discriminant is introduced into the framework of expected-model augmentation based on likely model-set(EMA-LMS).A kind of fuzzy inference-based EMA-LMS algorithm is proposed to adjust the parameter and strategy of model-set adaption online.This algorithm generates an expected model that is closer to the real motion model and makes better choices for model selection.The simulation results show that the proposed algorithm can strengthen the adaptive capacity for the uncertainty of target maneuver and measurement,and improve accuracy.

maneuvering target trackingvariable structure interacting multiple modelmaneuvering dis-criminantfuzzy inference

郝亮、黄颖浩、姚莉秀、蔡云泽

展开 >

上海交通大学自动化系

系统控制与信息处理教育部重点实验室,上海 200240

机动目标跟踪 变结构交互式多模型 机动判别 模糊推理

国家科技重大专项国防科技卓越青年科学基金

2018YFB13050032017-JCJQ-ZQ-031

2024

上海交通大学学报
上海交通大学

上海交通大学学报

CSTPCD北大核心
影响因子:0.555
ISSN:1008-7095
年,卷(期):2024.58(4)