Sensor and Computer Vision-Based Wheelchair Navigation System For Detecting Obstacle And Tactile Paving

Visual impairment has a significant impact, especially when accompanied by mobility impairment, resulting in loss of independent mobility. Many individuals with visual impairment rely on a white cane to navigate through tactile paving and to predict obstacles in their surroundings. However, using a white cane as a navigation aid for individuals with visual impairments who also use a wheelchair for mobility requires considerable effort to maneuver the wheelchair while holding the white cane. This study presents a wheelchair navigation system based on sensors and computer vision, equipped with obstacle and tactile path detection. The trained TensorFlow model achieved a total accuracy of 94.86% based on mAP50. The designed system demonstrates significant performance improvement with a completion time of 44.65 seconds, which is close to the normal condition time of 42.38 seconds. The NASA-TLX assessment demonstrates that the designed system can reduce the workload of using a wheelchair for visually impaired people than using a white cane.