166 A.R. Ortiz et al. Fig. 20.6 CoM’s Trajectory of the pedestrian: the dash line is the trajectory obtained in the experiment. The solid line is the trajectory of the model using the maximum probable values of the parameters (a) (b) Fig. 20.7 Pedestrian induced forces for the CoM’s trajectory: top view (a) and 3D view (b). The arrows are normalized so that the length of the longest is 1 solution for the identification of large trajectories. Additional work should be done to reduce frequency components included in the experimental data which are produced by the vibration of the car and the resolution of the MS Kinect. This research shows that there is a high correlation between two parameters of the model, which means that a less number of parameters could be used for modeling the trajectory of the pedestrian. The longitudinal and transversal components of the trajectory show good approximation respect to the maximum probabilistic values, however there is still high standard deviations in the probability density functions of parameters as the amplitude of the velocity, av, and the maximum longitudinal velocity, v0. This uncertainty is affecting the performance of the model. More tests, a better mobile platform, and a better definition of the part of the body used to identify the trajectory of the CoM could reduce the uncertainty in these parameters.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==