A Method for Identifying Laser-tracked Vehicles Using Velocity Data from V2V Messages

In recent years, since the performance of automotive devices such as range sensors, millimeter wave radars, and onboard cameras has been highly enhanced, it has become possible for a vehicle to acquire the distance to surrounding vehicles with high accuracy. Thus, some vehicle awareness techniques by utilizing these devices have been studied [1]. Such techniques are beneficial to detect the presence of surrounding vehicles and pedestrians autonomously although it is still challenging to recognize vehicles beyond their scanning range. On the other hand, a new paradigm has been considered for safe driving where the position information is shared with neighboring vehicles via V2V communication. Such cooperative vehicle awareness has yet to provide any positional relationship among multiple vehicles because many legacy vehicles do not equip any communication device in the transition period. However, in the early stage of automated vehicles deployment, it should be necessary for those vehicles to recognize the presence of legacy vehicles in a short time to assist quick and right decision making. In this work, we propose a method for identifying of surrounding vehicles detected by installed range sensor using V2V-based velocity information. Our proposed method assumes the transition period of connected vehicles on automation level 2, where the human driver performs part of the dynamic driving task, and level 3, where the automated driving system performs the entire dynamic driving task in accordance with NHTSA automated vehicle classifications [2]. Those vehicles share two types of information via V2V communication: 1) Own vehicle information such as position and velocity and 2) Surrounding vehicles’ positions detected by the installed range sensor.