Abstract
In this paper, we quantify the impacts of model fidelity on the effectiveness of trajectory optimization for autonomous vehicles when driving at the limits of friction through experiments with a full-size vehicle. Models ranging from a double-track model with lateral and longitudinal load transfer dynamics to a simple point-mass model are used in combination with direct numerical optimization to generate optimal trajectories subject to the limits imposed by each model. The effectiveness of each model for trajectory planning is evaluated by testing the trajectories on an automated vehicle across friction conditions ranging from ice to dry asphalt. Comparisons between the outright performance of the car and the car's ability to track the optimal trajectory are made across the various models. The tests reveal that the advantage of more complex models is less that they better predict the vehicle's behavior, but that they provide a more nuanced view of the vehicle's limits and guidance on the proper coordination of the various actuators on the vehicle in order to make most efficient use of the available tire friction.
Highlights
T HE development of increasingly capable autonomous vehicles and improvements in computer hardware and software for numerical optimization have cast optimal control of automobiles into the limelight
While trajectory optimization was once primarily interesting from an academic perspective [1], the ability to solve numerical optimal control problems rapidly has led to an explosion of optimal trajectory planning and model predictive control (MPC) schemes for autonomous vehicles
MPC has been used to achieve a level of vehicle control approaching that of a human being in situations ranging from highway driving and lane changing, to parking, to emergency obstacle avoidance, to minimum time maneuvering [2]–[5]
Summary
T HE development of increasingly capable autonomous vehicles and improvements in computer hardware and software for numerical optimization have cast optimal control of automobiles into the limelight. MPC has been used to achieve a level of vehicle control approaching that of a human being in situations ranging from highway driving and lane changing, to parking, to emergency obstacle avoidance, to minimum time maneuvering [2]–[5]. Emergency obstacle avoidance and minimum time maneuvering differ from the other tasks in that they require operation at the vehicle’s friction limits, albeit relatively briefly in the case of obstacle avoidance. This commonality means that insights from one field may be applied to the other and to any situation requiring operation at the limits. Date of publication January 15, 2021; date of current version August 23, 2021.
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call