Collaborative Robots in Geophysical Flows: Understanding How Local Measurements Imply Global Structures

(ONR YIP 2012-2015)

Geophysical fluid dynamics (GFD) is the study of natural large-scale fluid flows, such as oceans, the atmosphere, and rivers. GFD flows are naturally stochastic and aperiodic, yet exhibit coherent structure. Coherent structures are important because they enable the estimation of the underlying geophysical fluid dynamics. Which enables the prediction of various physical, chemical, and biological processes in GFD flows. Nevertheless, the data sets that describe GFD flows are often finite-time and of low resolution which limits the our ability to find and track coherent structures on such flows. The goals of this project are to overcome the theoretical and technical challenges of developing a general mathematical and control framework for distributed autonomous sensing and tracking of geophysical fluid dynamics. Specifically, the work will focus on the identification and tracking of a class of coherent structures that are important for quantifying transport phenomena in flows. The key idea exploits the capability of the team to cover large regions in physical space to increase the spatio-temporal sampling resolution of the flow field. The data will then be processed in a distributed fashion by the team to obtain a global description of the flow dynamics that can be maintained and updated in real time. The objectives of this proposal are to develop a new stochastic control framework for AUVs and ASVs to: 1) Collectively harvest coherent structure information on general flows using on-board sensing capabilities and other stationary sensors; 2) Control the spatial distribution of sensing resources to maximize information gain in time-varying and uncertain environments; and 3) Experimental validation of the proposed strategies via a proof-of-concept robotic system with simulated and realistic flow data.


  • M. A. Hsieh, E. Forgoston, T. W. Mather, and I. Schwartz, Robotic Manifold Tracking of Coherent Structures in Flows, accepted to the IEEE International Conference on Robotics and Automation (ICRA2012), May 2012, St. Paul, MN, USA. BibTeX | PDF

Ensemble Design of Resource-Aware Control Strategies for Multi-Agent Robotic Systems

(NSF 2011-2012)

The objective of this research is to generalize mean-field approaches from physics and chemistry for integrated design of scalable, network resource aware, distributed control strategies for multi-agent robotic systems. Mean-field theory has been used by physicists and chemists to study and analyze complex systems with multiple interacting components. At the microscopic level, these systems are composed of various stochastically interacting molecules whose individual behaviors are difficult to predict. At the macroscopic level, the dynamics of the ensemble statistics for these stochastic systems can be accurately modeled using mean-field theory. This idea is to generalize these mean-field methods to develop macroscopic models that retain salient features of the underlying multi-agent robotic system and use these models in the design of distributed control strategies.


  • T. W. Mather and M. A. Hsieh. Ensemble Modeling and Control for Congestion Management in Automated Warehouses, submitted to the IEEE International Conference on Automation Science and Engineering (CASE 2012), August 2012, Seoul, Korea. BibTeX | PDF
  • M. A. Hsieh, E. Forgoston, T. W. Mather, and I. Schwartz, Robotic Manifold Tracking of Coherent Structures in Flows, accepted to the IEEE International Conference on Robotics and Automation (ICRA2012), May 2012, St. Paul, MN, USA. BibTeX | PDF
  • T. W. Mather, M. A. Hsieh. Ensemble Synthesis of Distributed Control and Communication, accepted to the IEEE International Conference on Robotics and Automation (ICRA2012), May 2012, St. Paul, MN, USA. BibTeX | PDF


  • Distributed filtering for dynamic task allocation

Multi-Robot Systems for Large Scale Cooperative Tasks

(NSF 2011-2014)

The objectives of this project is to provide undergraduate and graduate students a unique opportunity to work with an interdiscplinary and international team of researchers on the design and control of multi-agent robotic systems. The technical focus of the collaboration is centered around the design of robust multi-robot coordination strategies for execution of large scale cooperative tasks. Advances in embedded processor and sensor technology in the last thirty years have accelerate the demand for teams of robots in various application domains. Multi-agent robotic systems are particularly well-suited to execute tasks that cover wide geographic ranges, require significant parallelization, and/or depend capabilities that are varied in both quantity and difficulty. Example applications include littoral exploration and surveillance, rainforest health monitoring, autonomous transportation systems, warehouse automation, and hazardous waste clean-up.

Overseas undergraduate and graduate co-op opportunities are available. Please email Dr. Ani Hsieh at mhsieh1 _at_ drexel dot edu to find out more.

Task Partitioning for Distributed Assembly

Description to come ...


  • J. Worcester and M. A. Hsieh. Task Partitioning via Ant Colony Optimization for Distributed Assembly, submitted to the ANTS 2012: 8th International Conference on Swarm Intelligence, September 2012, Brussels, Belgium.
  • J. Worcester, J. Rogoff, and M. A. Hsieh. Constrained Task Partitioning for Distributed Assembly, in the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS’11), Sept 2011, San Francisco, CA, USA. BibTeX | PDF
  • M. A. Hsieh and J. Rogoff. Complexity Measures for Distributed Assembly Tasks, in the Proc. of the 2010 Performance Metrics for Intelligent Systems Workshop (PerMIS’09), Baltimore, MD, USA, Sept 2010. BibTeX | PDF


  • Driving up to an assembly block
  • Placing an assembly block

Integrating Computational Thinking in Tomorrow's Mechanical Engineering Education

(Mathworks, Inc. 2010-2012)

The objective of this work is to develop a two-quarter Robotics course sequence focused on the fundamentals of scientific computating concepts for Mechanical Engineering students. The two-course sequence will consist of one sophomore /pre-junior level course and one junior/senior level course where students will be expected to implement solutions to problems using MATLAB. The first course will provide an introduction to computation and data analysis using MATLAB and cover concepts including variables, functions, flow control, iteration, and recursion. The second course will provide an introduction to data structures and algorithms and cover concepts including sorting, searching, and graph manipulations. This course will discuss a number of canonical problems and show how numerical methods are used to solve them. These concepts will be illustrated through Robotics and Automation related examples and assignments to help illustrate the role of computational thinking in various scientific and engineering problems. Examples and assignments will encompass the simulation of physical robotic systems, the analysis of experimental data, control of sensors and actuators, perception, and motion planning. Project outcomes and deliverables will include 1) a scientific computation curriculum centered around Robotics and Automation related applications designed for Mechanical Engineering students; 2) assignments developed around USARSim and SRV-1 robots to increase student awareness, familiarity, and comfort level with developing code in MATLAB; 3) increase students usage of MATLAB in other STEM related courses by introducing the proposed curriculum at the sophomore level; 4) MATLAB toolbox for interfacing with USARSim; 5) MATLAB toolbox for an overhead tracking system; and 6) updated MATLAB PointGrey Research libraries.