Autonomous path planning robot using geographical information

Abstract

The purpose of this project is initially to present an implementation of a guidance algorithm based on the geometrical analysis for the adoption of path planning problem. A mobile robot (named toward its unique characteristic in this project as constant speed robot) moving at a constant speed and having a constant swerving ability or no swerving is used in the analysis. The guidance algorithm is formed based on an initial direction and distance of the target location without considering the time taken by the robot to reach the desired destination. At the moment, the guidance algorithm managed to assist the robot to reach the desired destination, within a certain small error tolerance. The development of the path planning algorithm is initially presented by the geometrical analysis and it is then analyzed by simulation using Matlab. The simulation analysis is performed to prove the effectiveness and robustness of the guidance algorithm for the mobile robot. The performance of the guidance algorithm will be compared with an existing guidance algorithm based on kinematics geometry and with an alternative guidance strategy employing geometrical analysis and Matlab’s simulation. After the simulation using Matlab, the algorithm will then be implemented in the robot using basic stamp to run the mobile robot. A camera will be use to act as an ‘eye’ for the robot to avoid obstacle. After the robot had avoided the obstacle, the robot will then continue on its path to the targeted location. The accuracy is almost 80% during the experiments. The robot just went off its course by just 10°.