Wijaya, Dion Dwi
Unknown Affiliation

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

Analisis Penggunaan Sumber Daya Pada Jetson Nano Untuk Sistem Pengenalan Wajah Wijaya, Dion Dwi; Hugeng; Utama, Hadian Satria
Jurnal Terapan Teknologi Informasi Vol 8 No 2 (2024): Jurnal Terapan Teknologi Informasi
Publisher : Fakultas Teknologi Informasi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21460/jutei.2024.82.374

Abstract

Jetson Nano, a Single-Board Computer or SBC developed by NVIDIA, is used to implement a face recognition system that requires large computing resources. This research aims to analyze the resource usage on the Jetson Nano during the running of the face recognition system, including CPU usage, GPU, memory, and power consumption. Monitoring data was obtained using Jtop software to record real-time resource usage and a wattmeter for electrical power consumption. The results showed that the Jetson Nano CPU performed consistently with an average utilization of 77.14%, reflecting an even distribution of workload. The GPU showed an average utilization of 44.05% with higher fluctuations, indicating variations in graphics workload intensity. Memory was used close to the maximum capacity of 4 GB, with an average utilization of 3.75 GB, indicating efficient memory management. Average power consumption was recorded at 8.56 Wh, confirming the energy efficiency of this device. This study concludes that the Jetson Nano is capable of running the facial recognition system stably and efficiently, although there is room for further optimization on GPU load distribution and memory management. With its high power efficiency, the Jetson Nano is an ideal solution for artificial intelligence-based applications with low power requirements.
APLIKASI ALGORITMA SARSA DALAM PENGENDALIAN MOTOR DC Wijaya, Dion Dwi; Fat, Joni; Mawardi, Viny Christanti
Jurnal Informatika dan Teknik Elektro Terapan Vol. 13 No. 1 (2025)
Publisher : Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/jitet.v13i1.5770

Abstract

Motor DC telah menjadi komponen penting dalam berbagai aplikasi industri dan perangkat elektronik. Namun, kendala dalam kestabilan kecepatan akibat variasi tegangan dan beban membutuhkan solusi kontrol yang adaptif. Penelitian ini mengimplementasikan algoritma SARSA berbasis Reinforcement Learning untuk mengontrol parameter Proportional Integral Derivative secara dinamis. Algoritma ini dirancang untuk meningkatkan respons sistem dalam menghadapi perubahan kondisi operasional. Dataset diperoleh melalui simulasi MATLAB, kemudian digunakan untuk melatih tabel Q yang memandu keputusan algoritma SARSA. Hasil pengujian menunjukkan bahwa sistem mampu menyesuaikan parameter PID dengan adaptif, menjaga kecepatan motor mendekati nilai target meskipun terdapat fluktuasi pada fase awal dan transien. Rata-rata kesalahan sistem sebesar 12,96%, mengindikasikan ruang untuk optimasi lebih lanjut. Penelitian ini membuktikan efektivitas SARSA dalam meningkatkan kestabilan motor DC dan berpotensi diterapkan pada sistem kontrol lainnya yang memerlukan adaptasi real-time.