Multiple-Vehicle Cooperative Autonomous Parking Trajectory Planning Using Connected Fifth Degree Polynomials and Genetic Algorithm Optimization

Advancements in smart vehicle technologies have paved the way for collaborative driving maneuvers among multiple vehicles. Automated parking is a pivotal challenge during the early stages of fully automated vehicle deployment. This research deals with solving the challenge of multi-vehicle cooperative automatic parking. This innovative technique addresses the trajectory planning required for multi-vehicle cooperative parking scenarios. To tackle the complexities inherent in cooperative parking, a novel approach is introduced that utilizes the power of connected fifth-degree polynomials and genetic algorithms. This is achieved by establishing the vehicle mathematical model coupled with collision avoidance constraints. The collision free trajectory planning is then formulated as a new optimization problem, where the purpose is to minimize the parking time. The distinctive advantage of this approach is its capability for rapid computation and reliable optimization with a small number of decision variables along with minimizing the overall parking time by managing simultaneous cooperative parking. Because of efficient computational load during the optimization, this approach distinguishes itself from conventional methodologies. The research simulations evaluate the collaborative parking maneuver applied to an automated multi-vehicle system. Comprehensive simulations unequivocally demonstrate the remarkable efficacy of the proposed method in surmounting the challenges associated with automatic parking in a multi-vehicle cooperative setting.