Rahmadi, M. Agung Ghifariansyah (2026) Perancangan dan Simulasi Sistem Navigasi Otonom Quadcopter Berbasis Pengolahan Citra Visual dan Kendali SMC-Fuzzy untuk Inspeksi Jalur Pipa Migas. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5009211058-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5009211058-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (11MB) | Request a copy |
Abstract
Inspeksi jalur pipa migas secara manual memiliki keterbatasan dari segi efisiensi, keselamatan, dan kontinuitas pemantauan, terutama pada area yang sulit dijangkau atau berisiko. Oleh karena itu, penelitian ini merancang dan mensimulasikan sistem navigasi otonom quadcopter untuk inspeksi jalur pipa migas berbasis pengolahan citra visual dan kendali SMC-Fuzzy. Sistem computer vision digunakan untuk mendeteksi jalur pipa melalui segmentasi warna RGB, Canny edge detection, Hough transform, Region of Interest, dan validasi centerline. Hasil pengolahan citra menghasilkan parameter navigasi berupa elat, eang, pipe valid, vision class, turn hint, dan turn line found. Parameter tersebut digunakan oleh FSM Mission Generator untuk menentukan mode misi dan membentuk referensi gerak xd, yd, zd, dan ψd. Referensi gerak selanjutnya dikendalikan menggunakan Sliding Mode Control yang dikombinasikan dengan Fuzzy Logic Controller untuk meningkatkan kestabilan dan kehalusan respons quadcopter. Pengujian dilakukan pada lingkungan simulasi MATLAB/Simulink melalui skenario open loop, path following Misi 1, path following Misi 2, obstacle avoidance, disturbance visual, serta perbandingan SMC konvensional dan SMC-Fuzzy. Hasil simulasi menunjukkan bahwa sistem mampu mengikuti jalur pipa pada Misi 1 dengan RMSE lateral error sebesar 0,221, RMSE heading error sebesar 0,133 rad, dan settling time 3,55 s. Pada Misi 2 diperoleh RMSE lateral error sebesar 0,289, RMSE heading error sebesar 0,302 rad, dan settling time 18,4 s. Pada skenario obstacle avoidance dan disturbance visual, sistem memperoleh RMSE lateral error sebesar 0,276, RMSE heading error sebesar 0,194 rad, settling time 24,3 s, dan overshoot posisi 2,94 m. Dibandingkan SMC konvensional, SMC-Fuzzy menurunkan RMSE error roll sebesar 28,46%, pitch sebesar 21,55%, dan yaw sebesar 33,06%, serta menurunkan RMS sinyal kontrol U2, U3, dan U4 masing-masing sebesar 13,44%, 25,19%, dan 50,45%. Dengan demikian, sistem yang dirancang mampu mendukung navigasi visual quadcopter untuk inspeksi jalur pipa secara otonom, stabil, dan adaptif.
======================================================================================================================================
Manual oil and gas pipeline inspection has limitations in efficiency, safety, and monitoring continuity, especially in areas that are difficult to access or potentially hazardous. Therefore, this study designs and simulates an autonomous quadcopter navigation system for oil and gas pipeline inspection using visual image processing and SMC-Fuzzy control. The computer vision system is used to detect the pipeline path through RGB color segmentation, Canny edge detection, Hough transform, Region of Interest, and centerline validation. The image processing results generate navigation parameters, including elat, eang, pipe valid, vision class, turn hint, and turn line found. These parameters are then used by the FSM Mission Generator to determine the mission mode and generate motion references consisting of xd, yd, zd, and ψd. The generated references are controlled using Sliding Mode Control combined with a Fuzzy Logic Controller to improve the stability and smoothness of the quadcopter response. The system was tested in a MATLAB/Simulink simulation environment through open-loop testing, Mission 1 path following, Mission 2 path following, obstacle avoidance, visual disturbance, and a comparison between conventional SMC and SMC-Fuzzy. The simulation results show that the system was able to follow the pipeline path in Mission 1 with a lateral error RMSE of 0.221, heading error RMSE of 0.133 rad, and settling time of 3.55 s. In Mission 2, the system obtained a lateral error RMSE of 0.289, heading error RMSE of 0.302 rad, and settling time of 18.4 s. In the obstacle avoidance and visual disturbance scenario, the system achieved a lateral error RMSE of 0.276, heading error RMSE of 0.194 rad, settling time of 24.3 s, and position overshoot of 2.94 m. Compared with conventional SMC, SMC-Fuzzy reduced the RMSE of roll, pitch, and yaw errors by 28.46%, 21.55%, and 33.06%, respectively, and reduced the RMS of control signals U2, U3, and U4 by 13.44%, 25.19%, and 50.45%, respectively. Thus, the designed system can support autonomous, stable, and adaptive visual navigation of a quadcopter for pipeline inspection.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Climate Action, Computer vision, Finite State Machine, Quadcopter, SMC-Fuzzy |
| Subjects: | Q Science > QA Mathematics > QA9.64 Fuzzy logic T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing. T Technology > TJ Mechanical engineering and machinery > TJ217.2 Robust control T Technology > TJ Mechanical engineering and machinery > TJ222 Supervisory control systems. U Military Science > UG1242 Drone aircraft--Control systems. (unmanned vehicle) |
| Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis |
| Depositing User: | M. Agung Ghifariansyah Rahmadi |
| Date Deposited: | 04 Aug 2026 10:25 |
| Last Modified: | 04 Aug 2026 10:25 |
| URI: | http://repository.its.ac.id/id/eprint/143661 |
Actions (login required)
 |
View Item |