Rancang Bangun Lengan Robot Otonom Pemetik Cabai Menggunakan Deep Learning Terintegrasi dengan IoT

Ubaidillah, Ahmad (2025) Rancang Bangun Lengan Robot Otonom Pemetik Cabai Menggunakan Deep Learning Terintegrasi dengan IoT. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5009211105-Undergraduate_Thesis.pdf - Accepted Version Restricted to Registered users only Download (4MB) | Request a copy

Abstract

Penurunan jumlah petani dan rendahnya adopsi teknologi menyebabkan menurunnya efisiensi dan produktivitas pertanian, terutama dalam proses pemetikan cabai. Untuk menjawab tantangan ini, penelitian ini mengembangkan lengan robot otonom pemetik cabai berbasis deep learning dan Internet of Things (IoT). Sistem memanfaatkan model YOLOv11s untuk mendeteksi kematangan cabai dan kinematika balik untuk mengatur gerakan lengan robot. Antarmuka web juga dikembangkan untuk pemantauan kondisi robot secara real-time, dengan target akurasi deteksi >90%, tingkat keberhasilan pemetikan ≥60%, latensi <500 ms, dan packet loss <5%. Pengujian menunjukkan model deep learning mencapai mAP@0.50 sebesar 0,923 dan akurasi rata-rata 92%. Dalam simulasi posisi acak, keberhasilan pemetikan mencapai 75,6% dengan waktu rata-rata 8,9 detik per cabai, sementara pada uji keadaan nyata, tingkat keberhasilan mencapai 60%. Sistem pemantauan IoT berfungsi optimal dengan latensi rata-rata 100,28 ms dan tanpa kehilangan data (packet loss). Hambatan fisik seperti ranting dan keterbatasan jangkauan masih menjadi tantangan. Secara keseluruhan, sistem ini menunjukkan potensi besar dalam mengotomatisasi proses pemetikan cabai secara efektif dan efisien.
==================================================================================================================================
The declining number of farmers and the low adoption of technology have led to reduced efficiency and productivity in agriculture, particularly in the chili harvesting process. To address this challenge, this study developed an autonomous chili-picking robotic arm based on deep learning and the Internet of Things (IoT). The system employs the YOLOv11s model to detect chili ripeness and utilizes inverse kinematics to control the robotic arm's movements. A web-based interface was also developed for real-time monitoring of the robot’s status, with target performance including detection accuracy >90%, picking success rate ≥60%, latency <500 ms, and packet loss <5%. Testing showed that the deep learning model achieved a mAP@0.50 of 0.923 and an average detection accuracy of 92%. In randomized position simulations, the system achieved a picking success rate of 75.6% with an average picking time of 8.9 seconds per chili, while in real-world tests, the success rate reached 60%. The IoT-based monitoring system performed optimally, with an average latency of 100.28 ms and zero packet loss. Physical obstacles such as branches and limited reach remain challenges. Overall, the system demonstrates significant potential for automating the chili harvesting process effectively and efficiently.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	deep learning, IoT, kinematika balik, robot pemetik cabai, teknologi pertanian, agricultural technology, chili picking robot, deep learning, inverse kinematics
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing. T Technology > TJ Mechanical engineering and machinery > TJ211 Robotics. T Technology > TJ Mechanical engineering and machinery > TJ213 Automatic control.
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Ahmad Dzulfikar Ubaidillah
Date Deposited:	29 Jul 2025 01:13
Last Modified:	29 Jul 2025 01:13
URI:	http://repository.its.ac.id/id/eprint/121351

Actions (login required)

View Item