采用相对位置量测的紧耦合式多无人机协同定位算法

Tightly Coupled Cooperative Localization Algorithm for Multiple Unmanned Aerial Vehicles Based on Relative Position Measurements

针对仅使用惯性测量单元(IMU)的自主定位导航存在严重误差累积的问题,将IMU解算和置信传播(BP)算法进行紧耦合,在扩展卡尔曼滤波器(EKF)的框架下,使用无人机(UAV)间的相对位置量测直接更新大地坐标系下各无人机的位置状态,提出了一种适用于大地坐标系多无人机定位与导航的紧耦合的分布式协同定位算法(EKF-BP)。首先,实现了基于IMU量测和无人机间相对位置量测的集中式协同定位算法;接着,将BP算法扩展到大地坐标系,并与基于IMU的自主导航紧耦合,完成卫星拒止环境中多无人机系统的协同定位与导航;最后,将所提出的方法与基于协方差交叉的协同定位进行比较。仿真结果表明,分布式协同定位算法具有明显的计算效率优势,与基于协方差交叉的分布式协同定位方法相比,所提EKF-BP算法提高了多无人机系统在卫星拒止环境中的定位导航精度,是协方差交叉方法提升量的10倍左右。所提的分布式EKF-BP算法能够显著地抑制仅依赖IMU数据所导致的定位导航精度随时间迅速变差的问题。

To tackle the error-accumulation issue of autonomous localization and navigation utilizing inertial measurement unit(IMU),a tightly coupled approach that integrates IMU estimation and belief propagation(BP)algorithm within the framework of the extended Kalman filter(EKF)is proposed.The proposed EKF-BP method directly updates the position states of multiple unmanned aerial vehicles(UAV)in the geodetic coordinate system based on relative position measurements.Firstly,a centralized cooperative localization algorithm is implemented based on IMU measurements and relative position measurements among UAVs.Then,the BP algorithm is extended to the geodetic coordinate system and tightly coupled with IMU-based autonomous navigation to achieve cooperative localization and navigation of multiple UAV systems in satellite-denied environments.Finally,the proposed method is compared with the cooperative localization based on covariance intersection.The simulation results show that the proposed distributed cooperative localization method has an obvious advantage in computing efficiency as the number of UAVs increases.In comparison to the distributed cooperative localization based on covariance intersection,EKF-BP achieves enhanced precision in localization and navigation in satellite-denied environments,which is about 10 times the improvement of the covariance intersection method.The proposed EKF-BP method effectively suppresses the rapid accuracy degradation of dead reckoning relying merely on IMU measurements.