Abstract
With the increase in autonomous car technology, and advance driver assistance systems (ADAS), the demand for vehicle trajectory prediction is increasing. These systems mostly use many sensors such as lidar, radar, stereo cameras which are big and expensive to detect the location of the ego vehicles with respect to other vehicles. Such sensors are not readily available to Vulnerable Road Users (VRU) such as motorcyclist, cyclist as they mostly only rely on their phones or a small device for navigation and safety warnings. Trajectory prediction is important to VRU as it is able to predict if the trajectory of other vehicles is a threat or not. Most of the current Trajectory prediction for vehicles involves plotting a relative position of other vehicles with respect to the ego vehicle and this is not possible with just a mobile device. Hence, this paper proposes a method to predict trajectory of other vehicles based on detected vehicles center coordinates. This is achieved using Gaussian Process Auto-Regression based on past data of vehicle center coordinates to predict future coordinates on an x-y pixel plane using only a camera sensor and You Only Look Once (YOLO) vehicle detection.