基于虚拟噪声模型补偿AUV导航算法研究

Compensation for AUV's Navigation Algorithm Based on Virtual Noise Model

自主式水下机器人(AUV)同时定位与环境地图构建(SLAM)是实现水下未知环境自主导航的关键性技术,是机器人研究的热门课题之一。针对自主式水下机器人SLAM框架下应用扩展卡尔曼滤波(EKF)因模型线性化产生误差以及噪声统计未知的情形,采用一种基于虚拟噪声补偿技术的EKF算法,该方法可以把未知模型误差归入到虚拟噪声中去,运用噪声统计估值器在线估计噪声统计。以构建的AUV运动系统的模型为基准,从滤波精度、收敛性及算法稳定性方面,通过matlab仿真验证改进的EKF算法的效果。仿真结果表明,相对于传统的EKF算法,改进后的EKF算法估计精度更高,预期效果更好,有效提高了非线性滤波的性能。

The simultaneous localization and mapping（SLAM） of Autonomous underwater robot（AUV） is the key technology to realize the auto navigation for robot in the unknown environment of underwater, and it is one of the hot topics in the field of robotics research. In the framework of autonomous underwater robot SLAM, extended kalman filter（EKF） was applied to achieve the SLAM. For the model linearization errors and unknown noise statistics, EKF algorithm was used based on virtual noise compensation technology. This method could make the unknown model error into the virtual noise, and use the noise statistical to estimate the noise statistics. Constructing the AUV motion system model as a benchmark, the improved EKF algorithm was verified through matlab simulation from filtering accuracy, convergence and stability of the algorithm. The simulation results show that, compared with the traditional EKF algorithm, the improved EKF algorithm can get higher estimation precision, the expected effect is better, and can effectively improve the performance of nonlinear filtering.