Trajectory Tracking and Collision Avoidance on Smart Wheel Chair

Abstract

There have been many developments of wheelchairs as mobility aids, including electric wheelchairs. Wheelchairs sometimes still require manual steering. Therefore, in this research, a smart wheelchair is developed that can move automatically to the destination position from a predetermined position with a trajectory tracking system. The system deploys the odometry method, orientation angle using the output of the IMU-9DOF sensor with Kalman filter, and collision avoidance to avoid collisions with obstacles in front of it. The use of the Kalman filter improves the angle output that is close to the reference. In the trajectory tracking test, the wheelchair can approach the given reference position with a maximum error of 20 cm for the x-axis and 2 cm for the y-axis. The wheelchair collision avoidance test has been able to avoid collisions, as it successfully detects an obstacle less than 40 cm and avoids the collision correspondingly.