Experimental Test of a Real-time Collision-free Maneuver Generation Algorithm for Autonomous Driving

Описание: Experimental Autonomous Driving test carried out in Parma, Italy, on the campus area (urban roads) open to regular traffic. The main objective of the test is to demonstrate the efficacy of our algorithm (presented at the European Control Conference 2023) in generating real-time feasible maneuvers.

Abstact:
In this paper we propose a real-time maneuver generation algorithm for Autonomous Vehicles (AVs). Given a planar road geometry with static and moving obstacles along it, we are interested in finding collision-free maneuvers that satisfied the AV's dynamics and subject to physical and comfort limits. Based on longitudinal and transverse coordinates, we propose a novel collision avoidance constraint and formulate a suitable maneuver regulation optimal control problem. Maneuver regulation has intrinsic robustness with respect to standard trajectory tracking that stems from the requirement of following a desired path with a desired velocity profile assigned on it. The optimization problem is solved by using a nonlinear optimal control technique that generates (local) optimal trajectories. We demonstrate the efficacy of the proposed algorithm by providing numerical computations on two different scenarios. Finally, experimental results are presented to demonstrate the efficiency, both in terms of computational effort and dynamic features captured, of the proposed algorithm.