Parameters Tuning for Enhanced Automated Guided Vehicle Navigation in ROS/Gazebo Simulation Environment

Abstract

Automated Guided Vehicle (AGV) robot, a type of ground transportation vehicle that follows a predetermined path, is now in high demand in industrial operations and among researchers. AGV robot may improve it carrying capacity in the delivery operation through consistent and safe behaviour. The main challenge is in its navigation system when obstacles appear unexpectedly on its desired path, its limited abilities make it unable to avoid obstacles that would interfere with the smooth operation and decrease the quality of time. The aim of this research work is to present a tuning parameter of algorithms, namely the Pure Pursuit based on coordinates look-ahead distance for navigation and the Vector Field Histogram based on safety distance for avoiding obstacles. Robot Operating System (ROS) platform and Gazebo simulator environment are used to simulate the simulation testing for algorithms. According to the test results, the combination of these algorithms produced promising outcomes by demonstrating the AGV's capability to manoeuvre along a predetermined path, avoid obstacles, and return to its original path in order to reach its goal position.