Secured Public Transportation System using Ambient Intelligence

This research paper seeks to evaluate and analyze various ways of trajectory prediction models that accurately forecast either the short-term of 1-3 seconds or long-term of 3-5 seconds spatial coordinates of a given scene in order to acquire the trajectories of different road-agents. The model’s key responsibilities are to detect various road agents (object tracking) using sensors and to estimate their future trajectories, taking into account their past observations and other prospective actions. This concept seeks to prevent crashes and accidents as well as rapidly and efficiently reduce traffic congestion. Trajectory Prediction models are becoming more and more crucial for secure navigation in autonomous cars and traffic forecasting in traffic management systems.