| dc.contributor.advisor | Hardhienata, Medria Kusuma Dewi | |
| dc.contributor.advisor | Rahmawan, Hendra | |
| dc.contributor.author | Fitriani, Zahra | |
| dc.date.accessioned | 2024-08-16T00:18:22Z | |
| dc.date.available | 2024-08-16T00:18:22Z | |
| dc.date.issued | 2024 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/157539 | |
| dc.description.abstract | Pertumbuhan populasi global yang cepat menghadirkan tantangan serius terhadap ketahanan pangan, sementara lahan pertanian yang terbatas menuntut peningkatan efisiensi produksi. Agriculture 4.0, dengan fokus pada pertanian presisi, menawarkan solusi melalui integrasi teknologi robotika. Penelitian ini mengembangkan dan menguji algoritma RL Q-Learning dengan pendekatan konsep flow untuk merencanakan rute pergerakan robot pertanian di lingkungan nyata, dengan tujuan meminimumkan total state, jumlah belokan dan total waktu tempuh. Studi kasus yang diangkat dalam penelitian ini adalah kasus pengoptimalan rute robot pemanen buah di ATP IPB Cikarawang. Hasil pengujian menunjukkan bahwa algoritma modifikasi meskipun memiliki total state yang sama dengan algoritma pembanding, namun dapat menurunkan jumlah belokan dan jumlah waktu tempuh. Untuk kasus yang diujikan, algoritma yang diusulkan dapat mengurangi jumlah belokan sebesar 68.42% diikuti dengan penurunan total waktu tempuh sebesar 21.62%. | |
| dc.description.abstract | Rapid global population growth poses a serious challenge to food sustainability, while limited agricultural land demands increased production efficiency. Agriculture 4.0, with a focus on precision farming, offers solutions through the integration of robotic technology. The research developed and tested the RL Q-Learning algorithm with a flow concept approach to planning the movement route of agricultural robots in real environments, with the aim of minimizing the total state, the number of turns and the total time spent. The case study raised in this study is a case of optimization of the fruit harvesting robot route at the ATP IPB Cikarawang. The test results showed that the modified algorithm, despite having the same total state as the comparison, could decrease the number of spins and the amount of time spent. For the trial case, the proposed algorithm could reduce the count of turns by 68.42% followed by a reduction in the total time spent by 21.62%. | |
| dc.description.sponsorship | | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Modifikasi Algoritma Q-Learning untuk Pemilihan Jalur Optimal dalam Lingkungan Pertanian | id |
| dc.title.alternative | Modification of Q-Learning Algorithm for Optimum Path Selection in Agricultural Environments | |
| dc.type | Skripsi | |
| dc.subject.keyword | perencanaan jalur | id |
| dc.subject.keyword | konsep flow | id |
| dc.subject.keyword | Q-Learning | id |
| dc.subject.keyword | robot pertanian | id |
| dc.subject.keyword | flow concept | id |
| dc.subject.keyword | path planning | id |
| dc.subject.keyword | Q-Learning | id |
| dc.subject.keyword | agricultural robots | id |