基于改进最小二乘法的无人机轨迹预测研究
Research on UAV Trajectory Prediction Based on Improved Least Squares Method

高效精确地对无人机飞行轨迹进行预测是无人机领域的关键技术之一。针对现在无人机飞行轨迹预测方法的预测精度不足的问题，本文在进行大地坐标与空间直角坐标转换的基础上，提出一种基于最小二乘的无人机飞行轨迹预测算法，在无人机的飞行过程中，该算法能够不间断地更新轨迹数据并拟合飞行轨迹方程，从而达到对飞行轨迹的精确预测。仿真实验结果表明，所提出的轨迹预测算法误差可控制在0.5 m以内，相比较于BP神经网络预测模型，该算法的可靠性和精度更优。
Efficient and accurate prediction of UAV flight trajectory is one of the key technologies in the field of UAV. To address the problem of insufficient prediction accuracy of current UAV flight trajectory prediction methods, this paper proposes a least squares-based UAV flight trajectory prediction algorithm based on the conversion of geodetic coordinates and spatial Cartesian coordinates, which can uninterruptedly update the trajectory data and fit the flight trajectory equation during the flight of UAV, so as to achieve accurate prediction of flight trajectory. The simulation experimental results show that the error of the proposed trajectory prediction algorithm can be controlled within 0.5 m, and the reliability and accuracy of the algorithm is better compared with the BP neural network prediction model.