Paper Title
Speed Prediction for Autonomous Driving in Urban Ways

Abstract
Due to developments and needs of driverless scenarios in traffic various companies and research groups have developed their own vehicle and tested in urban ways. Today on field one can see many intelligent vehicles tasked with different objectives to execute like Google, Mercedes, Mitsubishi car... etc. All these vehicles have completed (driven) thousands of kms in intercity highways (expressway) without an accident they have caused but the other drivers. Although such development is already made, there is still no existing law how these vehicles should operate, and who/what should be in charge of these vehicles in front of the law. However, the need of further development of these vehicles is still waiting for us with its challenges. In recent years, the road safety, local traffic recognition, autonomous drive within urban ways has gained more importance since all these developed vehicles up to now mainly are operated on either in interurban ways or in presence of no real local traffic. Since safety is a milestone in autonomous driving for all tasks like collision avoidance, speed prediction, driving assistance, autonomous parking ...etc. risk assessment models concerning driving models become more essential for algorithms to be developed to control these vehicles. In terms of motion prediction, the existing algorithms can be classified into three groups: 1. Physics-based motion models, 2. Maneuver-based motion models, 3. Interaction-aware motion models. In autonomous driving when the model is chosen and prediction of motion is derived for future the risk should be defined with the evaluation of future motion of prediction. The term Risk itself brings a broad definition concept. The risk assessment concept can be classified into two groups: • Collision of future possible trajectories • Unavoidable collisions Index terms - Autonomous driving, speed prediction