Analysis of Prediction Models for Multi-robot System NMPFC

This paper analyzes the prediction model of a nonlinear model predictive formation controller (NMPFC) applied to control the formation of a team of omni directional mobile robots. The prediction model calculates future formation behaviors with respect to obstacles, team mates in formation, target, orientation, position in formation and control effort using a kinematic model to predict most of the formation terms. Nevertheless, the prediction of the robots in formation can be done by either dynamic models or pure kinematic models. Therefore, this paper presents an analysis in order to find a balanced robot's prediction model that stimulates the robots the converge in less time and minimizing the controller's cost function. Finally, results of experiments with simulated robots are presented and discussed.

[1]  Andreas Zell,et al.  A model-predictive approach to formation control of omnidirectional mobile robots , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[2]  Pedro Costa,et al.  MODELING OMNIDIRECTIONAL MOBILE ROBOTS: AN APPROACH USING SIMTWO , 2012 .

[3]  Moritz Diehl,et al.  NMPC based on Huber penalty functions to handle large deviations of quadrature states , 2013, 2013 American Control Conference.

[4]  Pedro U. Lima,et al.  Perception-driven multi-robot formation control , 2013, 2013 IEEE International Conference on Robotics and Automation.

[5]  Dimos V. Dimarogonas,et al.  Event-based model Predictive control for the cooperation of distributed agents , 2012, 2012 American Control Conference (ACC).

[6]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[7]  António Paulo Moreira,et al.  Multi-robot nonlinear model predictive formation control: Moving target and target absence , 2013, Robotics Auton. Syst..

[8]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[9]  Eduardo F. Camacho,et al.  Distributed model predictive control based on a cooperative game , 2011 .