Developed to model insect locomotion using a Lévy random walk, in conjunction with the tracking software Ubitrail and analysis package RUbitrail. Scripts to make similar random walks found here, and several variants of random walk functions found here
Figure 1. Sample 12000 frame trajectories, showing endogenous locomotion in the mealworm beetle, Tenebrio molitor (left), and a simulated trajectory produced by a correlated walk rule (right).
Figure 2. Correlated walks produced using various turning parameters, where rho = 0.1, 0.5, 0.9 and 0.994, from top left to bottom right.
Simulates X,Y-trajectories of football players using a Lévy random random walk, with realistic velocity, acceleration and pauses. Parameters fitted in this script using real player positions from a sample dataset used in this publication.
Figure 3. Left: Movement of one Tromsø IL player in the first minute (1200 frames) of the match vs. Strømsgodset (2013-11-03). Right: Simulated X,Y-trajectories for 1200 frames based upon parameters fitted to all players in this match. Background pitch created in gglot using drawPitch.R.