Abstract
This paper introduces a simulation for service robots based on Robot Operating System (ROS) and Gazebo. It aims to test and evaluate task planning algorithms. After building the environment and the service robot model, a 2D map is built. To make use of semantic navigation, we construct a topological map from the 2D map. Several functions are written to realize the robot’s necessary actions, including navigation, grasping, placing, handing over, and searching person/object. Answer Set Programming (ASP) is used for planning; due to its complicated coding and poor legibility, we define an easy to understand format and translate it into ASP. Finally, we test our task planning system within the uncertain Gazebo environment. The experimental results show the usability of the simulation for task planning tests.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
