Photonics Optics Tech Inc. will develop a algorithm entitled Higher Order Highly Dynamic Smart Trajectory Reconstructor (HOHD-STR or STR for short) to enable 1) The establishment of an online, intelligence based, dynamic and adaptable database for selecting error and measurement models for different segments of the trajectory; and 2) The self-correction and updating feature to incorporate improvements in previous models. In the training mode, the STR will be trained to mimic models produced/tunes by experts by using previous trajectories reconstructed by the expert (expert models) as a training data. To train STR we will use one of Artificial/Computational Intelligence methods including Neural Networks, Support Vector Machine, Case-based Reasoning, Decision Trees, Nearest Neighbors, Inductive Logic Programming or others that will be appropriate. The scope of this project is to investigate the feasibility of the proposed innovative HOHD-STR technique to reconstruct Highly Dynamic Trajectories that will increase accuracy and successfully address issues of on-line modeling, higher order terms, and non-synchronized sensor data, for arriving at an accurate and faster method of trajectory reconstruction.
Benefit: POT will develop the Higher Order Highly Dynamic Smart Trajectory Reconstructor (HOHD-STR) technology to enable the reconstruction of Highly Dynamic Trajectories that will increase accuracy and successfully address issues of on-line modeling, higher order terms, and non-synchronized sensor data. This new trajectory reconstruction algorithm will enable the achievement of an accurate and faster method of trajectory reconstruction that will benefit aerospace, avionic flight vehicle test and evaluation as well as for commercial flight monitoring and control.
Keywords: Highly Dynamic Vehicle, Nonlinear Kalman Filter Model, High-Order Highly Dynamic Smart Trajectory Reconstructor, Gps-Imu Sensors Trajectory Reconstruction
