Collaborative Research: An Adaptive Dynamic Programming Approach to the Coordination of Heterogeneous Robotic Sensors Networks

nsf_small

    NSF ECS 0925407

ONR_logo_small.jpg

September 3, 2010 – August 31, 2013

$225,536.00

NSF

PI:

S. Ferrari

Project Description

The research project aims to develop a novel adaptive dynamic programming (ADP) approach for planning and controlling sensor networks. The objective is to automate and optimize the high-level decisions, while simultaneously sending control commands to mobile sensors. The optimal strategy of planning mobile sensors can be applied to the Pacman game, where in both scenarios, agents are performing tasks like pursuing and evading. The solution can also be applied to robotics problems, such as maintaining surveillance by a pursuer of an evader in a world populated with obstacles.

Research Goals

  • Analyze and apply a networked hybrid dynamical system (NHDS) formalism
  • Develop ADP and Bayesian inference algorithms for hierarchical factored Markov decision models
  • Design discrete-event/continuous-state interface based on computational geometry

Journal Articles

  1. G. Foderaro, A. Swingler, S. Ferrari, “A Model-Based Approach to Optimizing Ms. Pac-Man Game Strategies in Real Time”, IEEE Trans. on Computational Intelligence and AI in Games, Vol. 9, No. 2, pp. 153-165, June 2017. [PDF]
  2. G. Foderaro, V. Raju, and S. Ferrari, “A Model-based Approximate λ-Policy Iteration Approach to Online Evasive Path Planning and the Video Game Ms. Pac-Man,” Journal of Control Theory and Applications, Vol. 9, No. 3, pp. 391-399, 2011.  [PDF]
  3. W. Lu, G. Zhang, S. Ferrari, M. Anderson, and R. Fierro “A particle-filter information potential method for tracking and monitoring maneuvering targets using a mobile sensor agent,” The Journal of Defense Modeling and Simulation: Applications, Methodologym Technology, June 2012.  [PDF]
  4. W. Lu, S. Ferrari, R. Fierro, and T. W. Wettergren, “Approximate dynamic programming recurrence relations for a hybrid optimal control problem.” SPIE Defense, Security, and Sensing. International Society for Optics and Photonics, 2012.  [PDF]
  5. N. Bezzo, R. Fierro, A. Swingler, and S. Ferrari, “Mobile Router Networks: A Disjunctive Programming Approach,” International Journal of Robotics and Automation, Vol. 26, No. 1, pp. 13-25, 2011.  [PDF]
  6. S. Ferrari, R. Fierro, B. Perteet, C. Cai, and K. C. Baumgartner, “A Geometric Optimization Approach to Detecting and Intercepting Dynamic Targets Using a Mobile Sensor Network,” SIAM Journal on Control and Optimization, Vol.48, No. 1, pp. 292-320, 2009.  [PDF]

Peer-Reviewed Conference Proceedings

  1. G. Foderaro, V. Raju, and S. Ferrari, “A Cell Decomposition Approach to Online Evasive Path Planning and the Video Game Ms. Pac-Man,” invited paper, Proc. IEEE Multi-Conference on Systems and Control (MSC), Denver, CO, September 2011.  [PDF]
  2. P. Cruz, R. Fierro, W. Lu, S. Ferrari, and T. A. Wettergren, “Maintaining Robust Connectivity in Heterogeneous Robotic Networks,” Proc. of SPIE, Conference on Unmanned Systems Technology, Session on Intelligent Behaviors, Baltimore, MD, April 2013.  [PDF]
  3. D. Tolic, R. Fierro, and S. Ferrari, “Optimal Self-Triggering for Nonlinear Systems via Approximate Dynamic Programming,” Proc. IEEE Multi-Conference on Systems and Control (MSC), IEEE International Conference on Control Applications (CCA), Dubrovnik, Croatia, October 2012, pp. 879-884.  [PDF]
  4. W. Lu, S. Ferrari, R. Fierro, and T.Wettergren, “Approximate Dynamic Programming (ADP) Recurrence Relationships for a Hybrid Optimal Control Problem,” invited paper, Proc. of SPIE, Vol. 8387 83870C-2, Unmanned Systems Technology XIII, Session on Intelligent Behaviors, Baltimore, MD, April 2012.  [PDF]
  5. S. Ferrari, M. Anderson, R. Fierro, and W. Lu, “Cooperative Navigation for Heterogeneous Autonomous Vehicles via Approximate Dynamic Programming,” invited paper, Proc. of the IEEE Conference on Decision and Control, Orlando, FL, December 2011, pp. 121-127.  [PDF]
  6. W. Lu, G. Zhang, S. Ferrari, R. Fierro, and I. Palunko, “An information potential approach for tracking and surveilling multiple moving targets using mobile sensor agents,” Proc. SPIE Conference, Unmanned Systems Technology XIII, Orlando, FL, 2011.  [PDF]
  7. S. Ferrari, R. Fierro, and D. Tolic, “A Geometric Optimization Approach to Tracking Maneuvering Using a Heterogeneous Mobile Sensor Network,” Proc. IEEE Conference on Decision and Control, Shanghai, China, December 2009.  [PDF]
  8. D. Tolic, R. Fierro, and S. Ferrari, “Cooperative multi-target tracking via hybrid modeling and geometric optimization,” Proc. Mediterranean Conference on Control and Automation (MED’09), Thessaloniki, Greece, January 2009, pp.440-445.  [PDF]
  9. R. Fierro, S. Ferrari, and C. Cai, “An Information-Driven Framework for Motion Planning in Robotic Sensor Networks: Complexity and Experiments,” Proc. IEEE Conference for Decision and Control, Cancun, Mexico, 2008, pp. 483-489.  [PDF]
  10. S. Ferrari, C. Cai, R. Fierro, and B. Perteet, “A Multi-Objective Optimization Approach to Detecting and Tracking Dynamic Targets in Pursuit-Evasion Games,” Proc. American Control Conference, New York, NY, 2007, pp. 5316-5321.  [PDF]

Presentations

  1. “On the Duality of the Robot and Sensor Path Planning,” Conference on Decision and Control, Firenze, Italy, December 2013. [PDF]
  2. “A Cell Decomposition Approach to Online Evasive Path Planning and the Video Game Ms. Pac-Man,” MSC Conference, Denver, Colorado, September 2011. [PDF]
  3. “An Information Potential Approach for Tracking and Surveilling Multiple Moving Targets using Mobile Sensor Agents,” SPIE Defense, Security, and Sensing Conference, Orlando, FL, April 2011. [PDF]
  4. “Optimal Control of Mobile Sensor Networks,” MAE Seminar Series, Department of Mechanical and Aerospace Engineering, Princeton University, November 2009.[PDF]