Скачать или смотреть Trajectory Modelling

03/25/2017. Successful implementation of Kalman filtering on raw ping pong ball position coordinates to obtain position and velocity estimates for each frame. The position and velocity estimates are then used to estimate a 3D trajectory for the ball from each processed frame. For each trajectory, an intersection with the end-of-table plane is found and statistical filtering is applied to remove outliers. The main plot shows filtered position data as hollow red circles, the resulting trajectory estimate as the solid black line, and the ball's projected intersection with the end of the table plane as a solid red dot after statistical filtering.