An Autonomous Multiple Robot Registration and Control System: Design Implementation and Performance Evaluation

Abstract

ROS is the most prominent middleware used by most researchers in robotic application development. Our research mainly depends on ROS technologies because most researchers currently work with ROS as middleware for many research projects. Controlling the robots through the Web interface is essential. Because in some instances, users may not be able to communicate with the robot directly because of some bad conditions in the environment where the robots are currently placed. Therefore, we have developed a Web interface to control all robots through the Internet. However, the ROS topics, nodes, and message formats used to subscribe and publish can differ from one robot to another when we work with multiple robots in the same environment. Therefore, when a user expresses high-level instructions through a Web interface, all multiple robots must understand instructions uniformly and take necessary actions accordingly without considering each robot’s internal software and hardware implementation. The first contribution of the research is to develop an algorithm to register all robots based on the main components of the ROS technology through the Web interface autonomously. The robot Registration Engine was developed with algorithms to complete the autonomous robot registration task. The second contribution is identifying the relevant ROS topics and nodes for each action when a user command gives through the Web interface. The ROS topic identification algorithm was developed successfully. The third contribution was to evaluate the system performance under different conditions and derive the equations for the delay in response time through the web interface, validating the equations derived. We have conducted several experiments to evaluate our system with delays in response time. The worst-case analysis was completed for all algorithms with Big O notation. Users and researchers can utilize Robot Registration Algorithm and ROS Topic Identification Algorithm to work with multiple robots through the Web interface. We have successfully implemented all algorithms in a simulated environment in Gazebo.