Automatic Parking Path Planning Based on Ant Colony Optimization and the Grid Method

Abstract

This paper analyzes the path planning problem in the automatic parking process, and studies a path planning method for automatic parking. The grid method and the ant colony optimization are combined to find the shortest path from the parking start point to the end point. The grid method is used to model the parking environment to simulate the actual parking space of automatic parking; then this paper makes some improvements to the basic ant colony optimization, finds the destination by setting the ants’ movement rules in the grid, and finds the shortest path after N iterations; since the optimal path found is a polyline, it will increase the difficulty of controlling vehicle path tracking and affect the accuracy of vehicle path tracking. The bezier curve is used to generate a smooth path suitable for vehicle walking. Finally, through matlab simulation, the obstacles in the environment are simulated, and the parking trajectory is obtained. The results show that the path planning method proposed in this paper is feasible.

Research on mobile robot path planning in complex environment based on DRQN algorithm

“…When the mobile robot arrives at the target point, the round ends and returns to the starting point for the next training. The updating formula of the Q-learning is shown in equation (10).…”

“…Path planning methods can be divided into two categories based on environment information, location point search, and constraints: conventional and reinforcement learning methods [8]. Conventional methods can be further divided into rule-based and heuristic search traditional methods, represented by A * and dynamic window algorithms [9]; graphical methods based on geometry and graph theory, represented by raster methods and Voronoi diagrams [10,11]; intelligent biomimetic algorithms based on the foraging and evolution of organisms, represented by swarms of bees, beetles, and sparrows [12]. The above methods have different advantages, but all have limited utilization of environmental information and poor path planning in unknown and dynamic environments.…”

Research on Parallel Parking of Distributed Drive Vehicle

“…The core part of automatic parking includes path planning and tracking, where path planning is mainly based on the path planning of search algorithms [1][2][3] and geometric methods [4][5][6] . Search algorithms are usually suitable for global path planning, where geometric planning algorithms calculate geometric constraints in local environments to obtain the target path more quickly and effectively [7] .…”

Research on parallel parking path planning based on the G-PSOA

“…In [5], it used RRT for parking PP however, the convergence speed using this algorithm was slow, and the planned path was poorly smoothed. In [6], it used the grid method ACO to find the optimal path. The optimal path was usually a broken line so it would increase the difficulty of vehicle tracking control.…”