自由空间激光通信APT(Acquisition,Pointing and Tracking)系统主要由粗跟踪系统和精跟踪系统组成,粗跟踪系统主要是完成对目标信号的捕获,以确保入射的信标光在精跟踪控制系统的动态范围内。介绍了粒子滤波在APT粗跟踪系统中的应用,针对粒子滤波在跟踪算法中存在的问题进行了改进,设计了颜色和轮廓双重信息融合的似然模型;为提高目标跟踪算法在异常情况下的有效性,采用了异常检测和恢复策略,并设计了跟踪光斑目标的模拟系统。实验表明:所提出的多信息融合的似然模型跟踪算法在目标尺度变化、背景干扰、姿态变化、部分遮挡等环境下均能稳定地跟踪目标,提高了大气激光通信链路的可靠性。
Acquisition,pointing,and tracking(APT) systems in free-space laser communication are mainly composed of coarse tracking system and fine tracking system.The acquisition of target signal is specifically completed by coarse tracking system to make sure beacon light lie in the dynamic range of fine tracking control system.In this paper,the application of the particle filter in coarse tracking system of APT was presented.Considering the existing problems on the particle filter introduced into the track algorithm,the likelihood model including color and outline dual information fusion was designed.In order to improve the effectivity of the track algorithm under abnormal circumstances,the anomalous detection and recover strategy were adopted.And then simulative system was implemented for tracking facula object.Experimental results show that the likelihood model of multiple information fusion can ensure object tracking even in background interference,posture variety,scale variety and partial occlusion environment.The stability of atmosphere laser communication link can be further improved.