Comparison of path planning in simulated robot

Ch’ng, Chee Yu’ng (2020) Comparison of path planning in simulated robot. Final Year Project, UTAR.

Preview

PDF Download (3955Kb) | Preview

Abstract

The development of technology is getting faster and faster, and the robotics revolution is also expanding rapidly. This is because robots can make a huge contribution to humans. The flexibility of robots has made it being capable of performing a diversity of tasks automatically (Benefits of Using Robotics, 2019). They can replace humans to do dangerous jobs, and complete the jobs more efficiently and accurately than humans. A basic task of a mobile robot is to move to a targeted point in order to perform the specified tasks. Hence, path planning algorithm such as A* algorithm and Dijkstra’s algorithm are essential for the robot to navigate efficiently from one place to the targeted place. However, different robots will work in different environments. If all the robots use the same path planning algorithm, it is possible that they may not bring the greatest benefit. This is because each path planning algorithm has its own applicable domain, performance, advantages and disadvantages in various situations. Thus, a comparison of path planning algorithm in simulated robot will be conducted in this study. There are 4 path planning algorithms will be compared, which are Dijkstra’s algorithm, A* algorithm, RapidExploring Random Tree (RRT) algorithm, and the last one is an algorithm modified from A*. The concept of the modified A* is to search the path in two directions. As a result of this paper, Dijkstra's algorithm suitable for finding the shortest distance. If you want to speed up the search time and there are fewer obstacles on the map, it is recommended to use A*, modified A* or RRT. However, if there are many obstacles on the map, RRT will not be suitable for searching. Robot Operating System (ROS) and Simple Two Dimensional Robot (STDR) simulator will be used in this project.

Actions (login required)