Intelligent Reconnaissance: Combining Sensor Fusion and Cooperative Control Strategies to Enable the Unmanned Reconnaissance Missions of the Future

Principal Investigators: Kira Barton, University of Michigan, bartonkl@umich.edu
Lauro Ojeda, University of Michigan, lojeda@umich.edu
Student: Michael Quann, University of Michigan
Government: Denise Rizzo, William Smith, U.S. Army TARDEC
Industry: Frank Koss, Andrew Dallas, SoarTech

Work begins in 2016.

Reconnaissance is a task that is particularly well suited for an intelligent vehicle system, as computers can observe a region for features of interest continuously using several methods of sensing. As the technologies are developed that will enable such a vehicle to unburden the occupants of certain reconnaissance tasks, there is a need for coordination and networking of these vehicles.

In this project, we will develop control strategies to coordinate a reconnaissance mission using a collection of unmanned ground and air vehicles in challenging terrain, where certain vehicles are capable of providing additional energy to the smaller assets.

This control architecture presents new management challenges given the specific combination of cooperative control and sensor data fusion management: (1) a new approach for cooperative control of a continuously evolving system will require new stability and robustness laws to ensure stable system behavior in the presence of variation and uncertainties; (2) the fusion of multiple sources of data from ground and/or aerial vehicles will require the development of a new data fusion algorithm that enables real-time decision making in the field; and (3) optimization techniques that utilize mission sensor data to update the terrain maps, and optimal strategies for conducting a reconnaissance mission under a variety of vehicle, terrain, and situational conditions.

References:

  • Quann, M., & Barton, K. (2015b, October). An Iterative Learning Control Approach to Multi-Agent Formations. In Dynamic Systems and Control Conference (DSCC), 2015. ASME.
  • Altın, B., & Barton, K. (2014). Robust iterative learning for high precision motion control through L1 adaptive feedback. Mechatronics, 24(6), 549-561.