Real-time automated modeling and control of self-assembling systems

We present the M3 framework, a formal and generic computational framework for modeling and controlling stochastic distributed systems of purely reactive robots in an automated and real-time fashion. Based on the trajectories of the robots, the framework builds up an internal microscopic representation of the system, which then serves as a blueprint of models at higher abstraction levels. These models are then calibrated using a Maximum Likelihood Estimation (MLE) algorithm. We illustrate the structure and performance of the framework by performing the online optimization of a bang-bang controller for the stochastic self-assembly of water-floating, magnetically latching, passive modules. The experimental results demonstrate that the generated models can successfully optimize the assembly of desired structures.

[1]  Hod Lipson,et al.  On-line assembly planning for stochastically reconfigurable systems , 2011, Int. J. Robotics Res..

[2]  E. Klavins,et al.  Programmable Self-Assembly , 2007, IEEE Control Systems.

[3]  Marco Dorigo,et al.  Self-Assembly at the Macroscopic Scale , 2008, Proceedings of the IEEE.

[4]  Alcherio Martinoli,et al.  Towards multi-level modeling of self-assembling intelligent micro-systems , 2009, AAMAS.

[5]  Nikolaus Correll,et al.  SwisTrack - a flexible open source tracking software for multi-agent systems , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[6]  Alcherio Martinoli,et al.  A trajectory-based calibration method for stochastic motion models , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[7]  Daniel T Gillespie,et al.  Stochastic simulation of chemical kinetics. , 2007, Annual review of physical chemistry.

[8]  Hod Lipson,et al.  Distilling Free-Form Natural Laws from Experimental Data , 2009, Science.

[9]  David H Gracias,et al.  Three-dimensional fabrication at small size scales. , 2010, Small.

[10]  Alcherio Martinoli,et al.  Modeling Self-Assembly Across Scales: The Unifying Perspective of Smart Minimal Particles , 2011, Micromachines.

[11]  Frank Schweitzer,et al.  Brownian Agents and Active Particles: Collective Dynamics in the Natural and Social Sciences , 2003 .

[12]  Ronald A. Howard,et al.  Dynamic Programming and Markov Processes , 1960 .

[13]  Michael D. Schmidt,et al.  Automated refinement and inference of analytical models for metabolic networks , 2011, Physical biology.

[14]  Eric Klavins,et al.  Hidden Markov Models for non-well-mixed reaction networks , 2009, 2009 American Control Conference.

[15]  Alan J. McKane.  Brownian Agents and Active Particles: Collective Dynamics in the Natural and Social Sciences , 2003 .

[16]  Alcherio Martinoli,et al.  Modeling Swarm Robotic Systems: a Case Study in Collaborative Distributed Manipulation , 2004, Int. J. Robotics Res..

[17]  Nikolaus Correll,et al.  Modeling and designing self-organized aggregation in a swarm of miniature robots , 2011, Int. J. Robotics Res..

[18]  Jehoshua Bruck,et al.  Programmability of Chemical Reaction Networks , 2009, Algorithmic Bioprocesses.

[19]  Dejan Milutinovic,et al.  Cells and Robots: Modeling and Control of Large-Size Agent Populations , 2007 .

[20]  Hod Lipson,et al.  Automated modeling of stochastic reactions with large measurement time-gaps , 2011, GECCO '11.

[21]  Spring Berman,et al.  Stochastic strategies for a swarm robotic assembly system , 2009, 2009 IEEE International Conference on Robotics and Automation.

[22]  Eric Klavins,et al.  Setpoint regulation for stochastically interacting robots , 2009, Auton. Robots.

[23]  Thomas A. Henzinger,et al.  The theory of hybrid automata , 1996, Proceedings 11th Annual IEEE Symposium on Logic in Computer Science.