Path generation and tracking for car automatic parking employing swarm algorithm

Abstract

This paper introduces a new technique for automatic column (parallel) parking of 4-wheeled cars. The intended technique is divided into two main stages. The first stage is the generation of the optimal path trajectory that the car follows so that it can be parked in a parking lot with the minimum acceptable dimensions, after the detection of the parking place. This stage is supported with a full parking scenario which is suitable for practical implementation. The second stage is the path tracking, which is realized through feedback control system based on four-input two-output fuzzy controller. The parameters of the fuzzy controller beside the required time for parking achievement are tuned by the Artificial Fish Swarm Algorithm (AFSA). The performance of the involved Swarm Based Fuzzy Controller (SBFC) is measured for solving the path tracking problem while operating on the car's kinematic model. The simulation results for the path tracking problem reflect the success of the intended automatic parking control system. © 2011 IEEE.