Abstract
In recent years, development of active safety technologies affects remarkable decrease of traffic fatalities, but the number of road accidents remains high. This paper focuses on an autonomous obstacle avoidance system as one of the active safety technologies to prevent crash accidents. There are two methods to avoid an obstacle which appears ahead suddenly, i.e. braking and steering. In terms of avoidance by the steering control, a method for calculating an optimal avoidance trajectory considering surroundings information relates to the lateral vehicle dynamics and its solution cannot be uniquely determined. And the avoidance by steering control is more effective for the forward obstacle avoidance situation when environment changes suddenly or at high-speed driving. Therefore this study proposes an autonomous obstacle avoidance system with combined braking and steering by using a micro electric vehicle. This paper focuses on a scenario where a frontal obstacle suddenly appears during driving in a path with walls in both sides of the vehicle. In the proposed system, the location of an obstacle and walls are acquired by a laser range finder. Then, an appropriate avoidance trajectory is derived geometrically by connecting two arcs from the surroundings information. In this process, different calculation methods are used for deriving the desired trajectory, depending on the distance between the obstacle and the walls. Next, the desired velocity and yaw rate to trace the derived avoidance trajectory are calculated. Then, the autonomous obstacle avoidance is accomplished by control of the driving torque and the front steering angle for the micro electric vehicle to follow the desired velocity and yaw rate. Its validity is verified by numerical simulations and experiments using the micro electric vehicle. The results indicate that the proposed system can be run in practice in real time.
Keywords: Preventive Safety, Collision Avoidance, Automatic Driving System, Motion Control, Electric Vehicle