Demo: Measuring Distance Traveled by an Object using WiFi-CSI and IMU Fusion

Abstract

Accurately measuring the distance traveled by an object or odometry, in indoor environments is important in many applications such as video-game controller tracking or robot route guidance. While the distance traveled by an object can be simply measured using an accelerometer, it is wellknown that distances measured with accelerometers suffer from large drift errors. In this paper, we demonstrate WIO, a WiFi-assisted Inertial Odometry technique that uses WiFi signals as an auxiliary source of information to correct such drift errors. The key intuition behind WIO is that, among multiple paths of a transmitted WiFi signal that arrive at a moving object equipped with a WiFi receiver, WIO can isolate the path that is most parallel to the object's direction of motion and use the change in the length of that path as an estimate of the traversed distance. WIO then fuses this distance estimate with the distance measured from an accelerometer on-board the object to correct drift errors. We implement WIO using commodity devices, and evaluate it on a robot car. Our results demonstrate an average error of just 4.37% in estimating the distance traversed by the car.