Intention Prediction Using Ground Based Sensors and Multiple Apriori Trajectory Estimation Models

Intention prediction for moving objects is a crucial component in several domains, especially in surveillance and security. Understanding the intention of the approaching object aids in critical decision making about the advancing object. Predicting the trajectory intention involves prior trajectory estimation and localization of tracked objects. The data obtained from various sensors provide a rich information repository from which such estimation may be inferred. Optical fiber sensors are increasingly utilized in trajectory es-timation algorithms to enhance localization accuracy for tracked objects. Their use involves leveraging distributed fiber optic sensing (DFOS) technologies such as Fiber Bragg Gratings (FBG) or Distributed Acoustic Sensing (DAS). These methods measure strain, vibration, and temperature changes along the fiber, providing precise spatial-temporal data. Advanced algorithms are leveraged to achieve precision tracking and reduce uncertainties in dynamic environments. In the present work, a novel algorithm is presented for the estimation and prediction of trajectories on a 2D grid system. The methodology is tested on simulated datasets and results are presented.