ROS-Based Global Path Planning for Autonomous Ground Robot Using the Pre-Built Map of the Environment

Abstract

Optimal global pathway planning has been a great area of research for many years and continues to develop new approaches. Without a suitable global path, a robot can get stuck in local minima. For path planning, a map is required which has been developed implying different sensors by moving the robot manually in an environment that is troublesome, costly, time taken to process, and lacks accuracy. This paper suggests a way to utilize pre-built maps of the environment. All the environments are built according to the structural map, we aim to use the pre-built map for path planning which is faraway accurate and detailed even to the extent of millimeters. We are introducing an approach that converts the pre-built map into a robot understandable format through image processing. The superiority and effectiveness of our approach have been verified using Robot Operating System (ROS) packages. ROS-supported path planning algorithms are applied on pre-built maps and results are compared with the result of the same approaches on sensor maps which produce better results in terms of pathway length and time taken.