Laser scanner-based prediction of pedestrian movements by filtering and classifying posture

Against the background of road safety, an algorithm is presented below that uses point clouds to make the most accurate prediction possible about the future position of pedestrians. A core element is to classify the current state of movement of pedestrians over a random forest. The focus is on early detection of changes between individual states. The classification is based on information about posture as a feature, which is obtained by analyzing the distance between the points of a cloud and the local planes it contains. The estimation of pedestrian position and dynamics using an Interacting Multiple Model Kalman Filter is the second core element. Finally, a prediction of the future position of a pedestrian is made using a combination of filter solution and classification. It is shown that the accuracy of this estimation exceeds the accuracy of a prediction based solely on the filter solution.