Real-time human walking estimation with radar

Radars can be used to observe persons. Animation of an observed human on the basis of Frequency Modulated Continuous Wave (FMCW) radar measurements in virtual reality considerably facilitates the interpretation of the radar measurements. These radar measurements give detailed information of the motion of the human body parts. The radar reflections are time-varying in range and Doppler. This information is used to animate a human in virtual reality. The delay between the observation and animation must be short for real-time applications. The paper describes a fast method for estimating human walking parameters from radar measurements and a simulated model. The parameters are estimated by minimizing the difference between a database with simulated model data and the real measurements. An approximation technique is used to estimate the essential walking parameters of the walking motion. The error with the exact solution is less than 10%. The computation time is less than the measurement time. This opens possibilities for real-time applications