Parallel Parking and Parking Assist System for Autonomous and Semi-Autonomous Vehicles

Modern autonomous vehicles are designed to provide convenience for people in their daily lives. A level 5 Advanced Driver Assistance System (ADAS) does not require any human attention and they are free from Geo-fencing. There is research going on and is not yet fully implemented. So this system can only be classified as fully autonomous driving level 1 to 4. Here Level-1 cars involves the human drivers and it is lowest level of automation. And in Level-2 there is no selfdriving but instead the human sits in the car and this level of vehicles controls both steering, accelerating/decelerating. In Level-3 includes human interaction, if the system designed fails, designed on the basis of environmental detection. Level-4 vehicles are Self-driving, there are speed limits already designed within the systems. Assistive parking is the first application of autonomous vehicles that is fully autonomous when considering both risks and environments. In this research, we propose parallel parking assistance for autonomous and semi-autonomous vehicles. The plan is based on the kinematic model of the vehicle which is used as a parameter for a Model Predictive Controller (MPC) to control the vehicle autonomously and use the A^* algorithm is a Informed Search Algorithm or a Best-First Search Algorithm it aims to determine the shortest path possible to reach the specified goal node, and here this algorithm is used to find the path planning and parking. Using this system, the vehicle is retrieved from the parking spot and re-parked by reversing along the obtained path. We propose methods of planning independent of a vehicle's initial position or orientation. The vehicle parks by following reference trajectories generated by the system.