Design and implementation of an EKF-based slam in a synchro-drive mobile robot using a laser scanner | |
| Author | Gutierrez, Maria Marinela Mariano |
| Call Number | AIT Thesis no.ISE-21-20 |
| Subject(s) | Mobile robots Robots--Control systems Kalman filtering Supervised learning (Machine learning) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Mechatronics |
| Publisher | Asian Institute of Technology |
| Abstract | Indoor mobile robots are robotic systems that have a certain level of autonomy. These vehicles have been studied thorough out the years for applications in the mapping and localization. Simultaneous localization and mapping (SLAM) is one of the underlying problems that concurrently estimates the map while identifying the pose estimate of the robot. This study proposed to design a synchronous drive mobile robot with independent steering and driving mechanisms guided by a chain/belt transmission Furthermore, the implementation of EKF-SLAM is addressed in this study using the information from the odometry and laser scanner. The EKF filter has two steps, the 1) prediction step and the 2) correction step. The predicted measurement is used to initially determine the state of the robot of the robot and the correction step uses the actual measurement for comparison with the prediction. In detail, the landmarks are extracted in the environment using point clustering. Clustered points less than 15 points reject these clusters which aren’t included in the observed landmarks. Also, a middleware called Robotic Operating System (ROS) is used to communicate the robot’s microcontroller to the computer and employ packages related to SLAM. The EKF SLAM method is evaluated through the calculation of the root mean square error between the predicted measurement (calculated EKF-SLAM data) and the observed measurement (actual measurement in the environment). It was found out that drifting of the chains contributed to the increase in the root mean square error of 0.2981 m (x axis) and 0.1589 m (y-axis). Also, the mean square error from the overall theoretical distance of 4.48 m gives the error of 2.70%. |
| Year | 2021 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Industrial Systems Engineering (ISE) |
| Chairperson(s) | Mongkol Ekpanyapong |
| Examination Committee(s) | Manukid Parnichkun;Pisut Koomsap |
| Scholarship Donor(s) | RG Battery & Parts Supply, Philippines |
| Degree | Thesis (M. Sc.) - Asian Institute of Technology, 2021 |