Abstract
This paper presents a new method to estimate 2D displacement based on the step-and-heading algorithm for three indoor motion modes: walking, jogging and running. Inertial measurement unit (IMU) is utilized for heading estimation based on indirect Kalman filter and for step detection as well. The IMU is placed at the lower back and an inverted pendulum is exploited to model the walking pattern, and then to compute the step length. Unlike previous models which are limited to walking only, an appropriate model is proposed to overcome the limitation and include scenarios like jogging and running. Experiments are conducted and the experimental results demonstrates the capability of the proposed algorithm with the overall averaged root mean squared error to be 2.6 meters for 40 meters experiments.