Data‐driven identification of group dynamics for motion prediction and control

Abstract

A distributed model structure for representing groups of coupled dynamic agents is proposed, and the least-squares method is used for fitting model parameters based on measured position data. The difference equation model embodies a minimalist approach, incorporating only factors essential to the movement and interaction of physical bodies. The model combines effects from an agent's inertia, interactions between agents, and interactions between each agent and its environment. Global positioning system tracking data were collected in field experiments from a group of 3 cows and a group of 10 cows over the course of several days using custom-designed, head-mounted sensor boxes. These data are used with the least-squares method to fit the model to the cow groups. The modeling technique is shown to capture overall characteristics of the group as well as attributes of individual group members. Applications to livestock management are described, and the potential for surveillance, prediction, and control of various kinds of groups of dynamic agents are suggested. C

Social Control of Herd Animals by Integration of Artificially Controlled Congeners

“…Cows were equipped with a small light-weight box for data collection during field experiments [8], see Figure 1. The box contains electronics for recording the GPS location of the animal as well as other sensor data (e.g.…”

“…Parts from these data were then used for calibrating the parameters for the model used in this paper. The calibration process as well as the animal experiments are described in more detail in [8]. For future experiments, the box provides a two-tier animal control system consisting of a set of speakers for applying arbitrary, differential sound stimuli and a set of electrodes that enable the application of differential electric shock.…”

“…In [8] we developed a linear-in-parameter model that provides good qualitative and quantitative agreement with various individual and collective metrics when used for simulating grazing cows. In this paper, we introduce a hybrid model, whose single discrete state variable corresponds to the behavioral modes grazing and stressed.…”

“…In this paper, we approach this goal for a cow herd by using a hybrid model of the herd dynamics. The model has been developed based on physical principles and calibrated using a system identification process based on data collected from 10 cows on the United States Department of Agriculture, Agricultural Research Services (USDA-ARS) experimental ranch in Las Cruces, New Mexico, USA [8].…”

“…This mechanism bears the potential for amplifying the effect of external stimuli provided to only a few cows. In this paper we extend the model from [8] to include these two social effects. The increased agreggation tendency of stressed animals is parameterized by a constant factor, and the propagation of stress from one animal to another is modeled by a radius of detection and a time constant allowing for temporal stress decay.…”