基于蒙特卡罗卡尔曼滤波的无人机目标定位方法
A UAV Target Localization Approach Based on Monte-Carlo Kalman Filter

为了增强小型无人机目标定位的精度和稳定性，提出一种基于蒙特卡罗卡尔曼滤波的目标定位算法。蒙特卡罗方法被用于估计卡尔曼滤波的初值，以及状态量和预测观测量的均值和协方差矩阵。定位过程中只使用无人机位置和激光测距值信息，利用卡尔曼滤波框架对目标位置进行递推。飞行实验表明，相比原有多点定位方法，新方法能够在满足实时性要求的前提下将精度由原来的20~30 m提高到10 m以内并实时给出误差估计，具备很大应用潜力。
To improve the accuracy and stability of the target localization via small UAV, a localization method based on Monte-Carlo Kalman Filter is proposed. The Monte-Carlo method is used to initialize the Kalman filter and to estimate the mean value and the covariance matrix of the state value and the predicted measurement value. The whole process only requires the location of the UAV and the distance between the UAV and the target measured by LASER equipment. The Kalman filter framework is applied to recursively estimate the location of the target. The flight experiments show that, with the real time constraint, comparing with the traditional multiple point localization method, the proposed approach improves the accuracy from approximately 20 to 30 meters to less than 10 meters, therefore has a considerably potential in future applications