Trough-based localization for autonomous agricultural robots | |
| Author | Piyanun Ruangurai |
| Call Number | AIT Diss. no.ISE-22-02 |
| Subject(s) | Autonomous robots Image processing Computer vision |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctoral of Engineering of Engineering in Mechatronics and Embedded Systems |
| Publisher | Asian Institute of Technology |
| Abstract | The navigation task is one of the most difficult tasks for an intelligent robot because of these three reasons: “Where am I?”, “Where am I going?” and “How do I get there.” For precision agriculture, high precision and accuracy equipments are required for path tracking but they are expensive. Thus, this research proposes to improve the localization method based on machine vision in order to obtain low cost and data richness. In this work, the direct seeding of rice tractor with a machine vision system is applied as the pivot study. During traveling along the paddy field, the nearby wheel track row is the reference for guidance and the control variables are the orientation and the lateral distance between a tractor and a nearby row. Three algorithms: Principal Components Analysis (PCA), Hough transform (HOUGH), and Random Sample Consensus (RANSAC) is used to estimate the trough and the performance comparison has been studied. The experimental results show that PCA and HOUGH methods are suitable to estimate the trough distance. A novel log-likelihood objective function is proposed to improve the accuracy of the trough estimate method. It provides the optimized trough position compared with the ground truth. Our optimization with the PCA method has the average errors between trough position and ground truth as 28.4 and 28.1 millimeters in two experiments in the year 2019/2 and 2020/1 respectively. |
| Year | 2022 |
| Type | Dissertation |
| 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) | Dailey, Matthew N.;Manukid Parnichkun |
| Scholarship Donor(s) | Rajamangala University of Technology Phra Nakhon,Thailand;Royal Thai Government;Asian Institute of Technology Fellowship |
| Degree | Thesis (Ph. D.) - Asian Institute of Technology, 2022 |