CAJ | 학술논문

针对转换瑞利滤波器( SRF)由于测角误差先验信息不准确而导致跟踪性能退化的问题,提出了一种基于交互式多模型转换瑞利滤波器( IMM-SRF)的目标跟踪方法。该方法采用多个不同测角噪声水平的SRF模型进行滤波,并且通过对滤波结果的自适应加权融合解决测角误差未知的问题。与SRF方法相比,该方法无需测角误差的先验信息,鲁棒性和实用性更好。仿真结果表明,IMM-SRF方法可实现对匀速运动和曲线运动目标的准确跟踪,并且收敛速度也较快,验证了所提方法的有效性。
침대전환서리려파기( SRF)유우측각오차선험신식불준학이도치근종성능퇴화적문제,제출료일충기우교호식다모형전환서리려파기( IMM-SRF)적목표근종방법。해방법채용다개불동측각조성수평적SRF모형진행려파,병차통과대려파결과적자괄응가권융합해결측각오차미지적문제。여SRF방법상비,해방법무수측각오차적선험신식,로봉성화실용성경호。방진결과표명,IMM-SRF방법가실현대균속운동화곡선운동목표적준학근종,병차수렴속도야교쾌,험증료소제방법적유효성。
Tracking performance degradation is usually caused owing to inexact prior information of angle measurement error for the shifted Rayleigh filter( SRF) method. To solve the problem,an interacting multi-ple model shifted Rayleigh filter( IMM-SRF) method is proposed for object tracking. It exploits multiple SRF models with different levels of angle measurement noise to filter,and then,performs adaptively weigh-ting fusion for the filtering results to solve the problem of unknown angle measurement error. Compared with the SRF method,the IMM-SRF method has better robustness and practicality due to the fact that it does not require prior information of angle measurement error. Simulation results show that the IMM-SRF meth-od can track the uniform motion and curve motion objects with fast convergence speed,which verifies the effectiveness of the proposed method.