A Novel Current Estimation and Cooperative Localization Method Under Unknown Current and Non-Gaussian Noise

In this article, we uniquely investigate a multi-autonomous underwater vehicles (AUVs) cooperative localization (CL) method that simultaneously tackles ocean current velocity interference and complex non-Gaussian noise, which can better cope with practical engineering problems. First, aiming at the situation that the underwater Doppler velocity log (DVL) cannot track the bottom and the measurement noise is interfered with by outliers, this article constructs a new CL framework that takes into account the interference of ocean currents and the measurement outliers. Second, considering that the existing entropy-based filtering usually adopts a fixed-center Gaussian kernel function, which is difficult to cope with the complicated and changeable ocean environment, and based on the fact that the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\ell _{1}$ </tex-math></inline-formula>-norm measure is more robust to outliers than the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\ell _{2}$ </tex-math></inline-formula>-norm, we propose an improved variable-center maximum entropy filtering with the Laplacian as the kernel function. Finally, since the kernel width is a key parameter affecting the performance of the algorithm, we adopt an improved iterative method to compute the free parameters, which improves the flexibility of the algorithm. The effectiveness and superiority of the proposed method were verified by a multigroup comparison test.