To improve the stability and accuracy of manoeuvring target tracking in three-dimensional space based on the angle of arrival (AOA) and its rate of change observations, this study presents a new observation fusion method by fusing the received signal strength (RSS) with AOA and the rate of change of AOA. To enhance the adaptive ability of traditional strong tracking extended particle filter (TSTEPF) against model mismatch, this study re-determines the position of the fading factor in the strong tracking extended Kalman filter based on the orthogonal principle and gives the calculating method. And by combining the method with the particle filter, a new strong tracking extended particle filter (STEPF) algorithm is proposed. Simulation results show that after fusing RSS into the observation model, the tracking speed and precision are both improved, especially precision, as the position root-mean-square error has a 58% decline on average. And it is found that STEPF proposed in this study has a more stable adaptive ability than TSTEPF, and is superior in terms of position, velocity, and acceleration estimation accuracy.

中文翻译：

---

强力跟踪扩展粒子滤波器，用于机动目标跟踪

为了提高基于到达角（AOA）及其变化率观测值的三维空间机动目标跟踪的稳定性和准确性，本研究提出了一种通过将接收信号强度（RSS）与AOA融合的新观测融合方法和AOA的变化率。为了提高传统的强跟踪扩展粒子滤波器（TSTEPF）对模型不匹配的适应能力，本研究基于正交原理重新确定了衰落因子在强跟踪扩展卡尔曼滤波器中的位置，并给出了计算方法。并结合该方法和粒子滤波算法，提出了一种新的强跟踪扩展粒子滤波算法。仿真结果表明，将RSS融合到观测模型后，跟踪速度和精度均得到改善，尤其是精度，因为位置均方根误差平均下降了58％。并且发现，本研究提出的STEPF比TSTEPF具有更稳定的自适应能力，并且在位置，速度和加速度估计精度方面都优越。