Coarse alignment is the process where the initial transformation matrix between body and navigation frame is determined using inertial sensors output. For low-cost inertial sensors only the accelerometers readings are processed to yield the initial roll and pitch angles. , Since inertial navigation systems (INS) requires the attitude initial condition prior to its operation, the coarse alignment process must be made before. The accuracy of the coarse alignment procedure is vitally important for the navigation solution accuracy due to the navigation solution drift accumulating over time.
In this paper, instead of using traditional approaches, we propose using machine learning (ML) approaches to conduct the coarse alignment procedure. To that end, a new methodology for the alignment process is proposed, based on state-of-the art ML approaches, including Random Forests (RF) and more advanced boosting methods such as gradient tree XGBoost and Light Boost. Results from simulated alignment of stationary INS scenarios are presented. The results are compared with the traditional coarse alignment methods in terms of time to convergence and accuracy performance. Using the proposed approach, results show a significance improvement of the accuracy and time required for alignment process of an INS for autonomous platforms. The results show that RF performs consistently good followed by boosted trees. The contribution of the proposed approach to CA performance will enable enhanced performance of autonomous platforms operation.