AI-based smart and intelligent wheelchair

Abstract

The differently abled and/or old-aged people require assistance for their movement. Generally, such assistant providing tool is wheelchair. Normal wheelchairs are manually operated and heavy to move adding burden to the suffered. Hence, automated wheelchairs that are equipped with sensors and a data processing unit constitute a special class of wheeled mobile robots, termed as “smart wheelchairs” in general. In the existing system, the wheelchair movement that is controlled by joystick uses buttons to start and stop the wheel. This is difficult for the differently abled to press the required button with precision. Although there are smart wheelchairs with gesture control, it lacks accuracy in the calculation of the location. The proposed system uses artificial intelligence for its working and proves to be a unique combination of wheelchair and health monitoring system. The wheelchair can be accessed both in manual and automatic modes. In the manual mode, the wheel is controlled using joystick whereas in the automated mode, MPU6050 sensor and accelerometer is used to control the direction by gesture. SPO2 sensor attached to the wheelchair is used to collect the health parameters. Thus, enabling the self-dependency of the person. Further, deep learning analysis of the data from the sensors and the wheelchair usage pattern is compared with the dataset to determine the stress level. The signal from the sensors is monitored and the vitals data is updated in the ThingSpeak website via Bluetooth module serving as a digital health chart.