Sistem Kendali Manuver Kursi Roda Elektrik Berbasis Gestur Tangan Menggunakan Sensor Leap Motion

Rahmawati, Maulida Ika (2026) Sistem Kendali Manuver Kursi Roda Elektrik Berbasis Gestur Tangan Menggunakan Sensor Leap Motion. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5023221015-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Keterbatasan fungsi motorik pada lansia dapat menyebabkan kesulitan dalam mengoperasikan kursi roda elektrik berbasis joystick karena membutuhkan koordinasi dan kontrol tangan yang presisi. Berbagai metode kendali alternatif berbasis sinyal biomedis seperti EEG, EMG, dan EOG telah dikembangkan, namun penggunaan sensor yang terpasang langsung pada tubuh dapat mengurangi kenyamanan pengguna dalam penggunaan jangka panjang. Oleh karena itu, penelitian ini mengembangkan sistem kendali kursi roda elektrik berbasis gestur tangan menggunakan sensor Leap Motion sebagai metode kendali non-kontak. Sistem memanfaatkan data orientasi tangan berupa sudut pitch dan roll yang diproses menggunakan metode threshold untuk menghasilkan perintah navigasi berupa maju, mundur, belok kanan, belok kiri, zona netral, dan berhenti. Perintah kemudian dikirim melalui komunikasi Bluetooth ke mikrokontroler STM32F103C8T6 untuk mengendalikan aktuator motor DC secara real-time. Sistem juga menerapkan mekanisme one-shot gesture activation, zona netral, serta prioritas berhenti sebagai fitur keselamatan. Hasil pengujian menunjukkan bahwa Leap Motion memiliki jarak deteksi efektif pada rentang 20–30 cm dengan tingkat keberhasilan deteksi gestur sebesar 100% pada kondisi indoor. Pada pengujian pengguna, seluruh subjek berhasil menyelesaikan lintasan dengan tingkat keberhasilan 100%. Tingkat keberhasilan gestur mencapai 95,96% pada kelompok normal tanpa sandaran tangan, meningkat menjadi 98,2% setelah penambahan sandaran tangan, dan mencapai 92,26% pada kelompok lansia. Evaluasi usabilitas menunjukkan peningkatan indeks dari 79,93% menjadi 88,4% setelah pengembangan ergonomi, sedangkan kelompok lansia memperoleh nilai sebesar 92,8%. Pengujian ulang terhadap empat subjek lansia setelah jeda waktu satu bulan menunjukkan bahwa waktu tempuh rata-rata subjek mengalami peningkatan performa, yaitu dari 70,7 detik menjadi 38,86 detik. Berdasarkan hasil tersebut, sistem kendali kursi roda berbasis gestur menggunakan sensor Leap Motion mampu diterapkan sebagai metode kendali non-kontak yang intuitif, nyaman, dan aman untuk mendukung mobilitas pengguna pada lingkungan indoor.
======================================================================================================================================
Motor function limitations in the elderly can cause difficulties in operating joystick-based electric wheelchairs because they require precise hand coordination and control. Various alternative control methods based on biomedical signals such as EEG, EMG, and EOG have been developed, but the use of sensors attached directly to the body can reduce user comfort in long-term use. Therefore, this study developed a hand gesture-based electric wheelchair control system using the Leap Motion sensor as a non-contact control method. The system utilizes hand orientation data in the form of pitch and roll angles processed using the threshold method to generate navigation commands such as forward, backward, turn right, turn left, neutral zone, and stop. The commands are then sent via Bluetooth communication to the STM32F103C8T6 microcontroller to control the DC motor actuator in real-time. The system also implements a one-shot gesture activation mechanism, neutral zone, and priority stop as safety features. Test results show that Leap Motion has an effective detection distance in the range of 20–30 cm with a gesture detection success rate of 100% in indoor conditions. In user testing, all subjects successfully completed the trajectory with a 100% success rate. The gesture success rate reached 95.96% in the normal group without armrests, increased to 98.2% after the addition of armrests, and reached 92.26% in the elderly group. The usability evaluation showed an increase in the index from 79.93% to 88.4% after the ergonomics development, while the elderly group obtained a score of 92.8%. Retesting on four elderly subjects after a one-month interval showed that the average travel time of the subjects experienced an increase in performance, namely from 70.7 seconds to 38.86 seconds. Based on these results, the gesture-based wheelchair control system using Leap Motion sensors can be implemented as an intuitive, comfortable, and safe non-contact control method to support user mobility in indoor environments.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	kursi roda elektrik, sensor Leap Motion, gesture recognition, elderly, teknologi asistif,electric wheelchair, Leap Motion sensors, gesture recognition, elderly, assistive technology
Subjects:	R Medicine > RM Therapeutics. Pharmacology > RM950 Rehabilitation technology.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Maulida Ika Rahmawati
Date Deposited:	01 Aug 2026 03:36
Last Modified:	01 Aug 2026 03:36
URI:	http://repository.its.ac.id/id/eprint/141213

Actions (login required)

View Item