FAU Center for Connected Autonomy and AI

Overview

Simultaneous Localization and Mapping (SLAM) is a crucial technique in the field of robotics, as it enables a robot to map its surroundings and determine its location simultaneously. The widely used open-source Gmapping package in ROS is not the most accurate method for SLAM in autonomy. In this project, we will compare the accuracy of Gmapping and the Clearpath Dingo-O ground robots packaged with the proprietary Autonomy Research Kit (ARK) software for robust and efficient SLAM in real-world deployments. 

The purpose of this project is to compare the accuracy of Gmapping and ARK for SLAM in real-world deployment, which will allow us to determine which software will be more efficient for multi-robot SLAM. By gathering two different types of location data from Clearpath Dingo-O robots, we will be able to analyze the accuracy and effectiveness of Gmapping and ARK in regard to autonomous driving. The data collected should provide a clear indication of which software is better suited for autonomous driving,

Community Benefit

Multiple robots can complete exploration and mapping tasks in less time, and more accurately, than a single robot can. Cooperative robots stand to bolster many essential functions in society, from government applications to regular household use. Tasks that were previously dependent on human effort, and as a result, human limitations, will be able to be performed in a more efficient, cost-effective manner, often requiring less materials.

Team Members

• Jordan Chung-Lee
• Thiago Goncalves Vasconcelos
• Jonathan Mercer 
• Wendy Paillant
• Anne Telcy

Sponsors

FAU Center for Connected Autonomy and AI