Self-adaptive collaborative navigation and filtering method

The invention relates to the field of collaborative navigation algorithms, in particular to a self-adaptive collaborative navigation and filtering method used under unknown measurement noise. The method includes the steps of interaction process input, model filtering, model probability update and interaction output. Compared with a traditional collaborative navigation and filtering method, under the condition of unknown noise environments, the method can effectively reduce the problem of precision reduction caused by overlarge noise preset deviation. The method has a very good modular characteristic, people can increase the number of models and can also freely select filters with various characteristics, and accordingly the method has very good later correction performance. 本发明涉及协同导航算法领域，具体涉及种针对未知量测噪声下的具有自适应的协同导航滤波方法。本发明包括：输入交互过程；模型滤波；模型概率更新；交互输出。相比于传统的协同导航滤波方法，在噪声未知的环境下条件下，本方能有效减小因为噪声预设偏差过大导致的精度下降问题。本方法具有很好的模块化特性，所以我们不仅可以增加模型数量，而且还可以自由的选取具有各种特性的滤波器，因此其具有很好的后期修正性。