A Multi-Objective Optimization-based Path Planning Method for Parallel Parking of Autonomous Vehicle via Nonlinear Programming

In recent years, Autonomous Vehicle has become highly desirable to improve efficiency of transportation, to reduce number of accidents and to reduce travelling cost. Among the common tasks in the autonomous driving system, parallel parking is one of the most important tasks, which is performed very frequently as a daily routine. Thus, planning an efficient path for parallel parking significantly helps to reduce the cost and improve the efficiency, which is of great interests at both academia and industry. In this paper, we propose a multi-objective optimization formulation and develop a Nonlinear Programming based method for the path planning problem of the parallel parking task. The proposed method is demonstrated to be able to solve the path planning problem for parallel parking efficiently and robustly with good optimization results as well as convergence property in the computational studies. We also conduct several analysis of the optimization algorithm to explain the impacts of the environmental parameters and the objectives in the multi-objective function.