Moving a Robot in Unknown Area Using Robot Operating System and Gazebo Simulator

Abstract

Nowadays robots have become a most important asset. With the development of technology, we can do so many things using modern robotic technology. Normally, we can see most of the robots can identify a known area. From this study proposes the mechanism to move a robot in an unknown area. This has become a most important fact in day to day life. Because we can use this kind of robot in hazardous environments, and we can use this robot in several ways. The unknown environment in this sense, does not have an identified map. In that case, robots should have the capability to make decisions dynamically. When we implement this system, the robot needs to use the Artificial Intelligent (AI) Knowledge and decide whether there is a chance to move forward or not. The AI process is just like the heart of this robot. Because the robot has to identify if there is an object or not, check whether there is a chance to move forward or not, how the robot can move, does the robot should by-pass something without colliding, and does the robot should by-pass something, those kinds of decision-making part happen using this AI Process. The proposed system is based on the Robotic Operation System (ROS) and the simulator, which is going to be used is Gazebo. This project hopes to detect the object using a laser sensor or 2D LiDAR or vision base. Detected object avoidance part also can be done using the laser sensor 2D LiDAR sensor. After that, there should be a direction changing mechanism in the developing algorithm. That can be implemented with looping. Because after the robot changes direction it again needs to check whether another object is there in the navigated location. After that algorithm should have a proper navigation mechanism. Just moving the robot, we can not get an idea about the area in which we can do the movement safely. In that case, in the proposed system, I will generate a map using Simultaneous localization and mapping (SLAM). To do the mapping part, I need to use the Rviz application. That map will help to identify the areas which can do the movements without having any crashes. Another important function of this research study is reverse navigation. If in any case, the road ends up, then the robot will be able to do the reverse navigation and continue its journey. In this proposed algorithm I have written a method to do the reverse navigation. This study will be useful to have autonomous navigation in unknown areas.