No abstract

No abstract

No abstract

This paper demonstrates that an autonomous vehicle can perform emergency lane changes up to the friction limits through real-time generation and evaluation of bi-elementary paths. Path curvature and friction determine the maximum possible speed along the path and, consequently, the feasibility of the path. This approach incorporates both steering inputs and changes in speed during the maneuver. As a result, varying path parameters and observing the maximum possible entry speed of resulting paths give insight about when and to what extent a vehicle should brake and turn during emergency lane change maneuvers. Tests on an autonomous vehicle validate this approach for lane changes near the limits of friction.

Active safety systems enabled by steer-by-wire technology can share control with a driver, augmenting the driver's steering commands to avoid collisions and prevent loss of control. The extent to which this can be done is limited by the controller's ability to anticipate dangerous scenarios in order to appropriately intervene and steer the vehicle to safety. However, the non-linear nature of tire dynamics poses a challenge in predicting and modifying vehicle behavior in real-time. In this paper, online successive linearizations of the future planned vehicle trajectory approximates these non-linear dynamics in a real-time, model predictive controller that shares control with a human driver. Simulation results of aggressive maneuvers demonstrate the usefulness of this approach as well as illustrate interesting interactions between the sometimes competing objectives of vehicle stability and collision avoidance.