arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Human-Computer Interaction

Electric wheelchair navigation based on hand gestures prediction using the k-Nearest Neighbor method

Khairul Anam, Safri Nahela, Muchamad Arif Hana Sasono, Naufal Ainur Rizal, Aviq Nurdiansyah Putra, Bambang Wahono, Yanuandri Putrasari, Muhammad Khristamto Aditya Wardana, Taufik Ibnu Salim

· Mechatronics Electrical Power and Vehicular Technology

See on the map

Abstract

The advancement of technology in the medical field has led to innovations in assistive devices, including wheelchairs, to enhance the mobility and independence of individuals with disabilities. This study investigates the use of electromyography (EMG) signals from hand muscles to control a wheelchair using the k-Nearest Neighbor (kNN) classification method. kNN is a classification algorithm that identifies objects based on the proximity of similar objects in the feature space. The wheelchair control process begins with the development of a kNN model trained on EMG signal data collected from five respondents over 30 seconds. The data was processed using feature extraction techniques, namely Mean Absolute Value (MAV) and Root Mean Square (RMS), to identify motion characteristics corresponding to five types of movement: forward, backward, right, left, and stop. The extracted features were classified using the kNN algorithm implemented on a Raspberry Pi 3. The classification results were then used to control the wheelchair through an Arduino UNO microcontroller connected to a BTS7960 motor driver. The study achieved an average accuracy of 96% with the MAV feature and 𝑘 = 3. Furthermore, combining MAV and RMS features significantly improved classification accuracy. The highest accuracy was obtained using the combination of MAV and RMS features with 𝑘 = 3, demonstrating the effectiveness of feature selection and parameter tuning in enhancing the system's performance.

More like this