Optimized routing of unmanned aerial systems to address informational gaps in counterinsurgency
Author(s)Lee, Andrew C. (Andrew Choong hon)
Massachusetts Institute of Technology. Dept. of Civil and Environmental Engineering.
John M. Irvine and Cynthia Barnhart.
MetadataShow full item record
Recent military conflicts reveal that the ability to assess and improve the health of a society contributes more to a successful counterinsurgency (COIN) than direct military engagement. In COIN, a military commander requires maximum situational awareness not only with regard to the enemy but also to the status of logistical support concerning civil security operations, governance, essential services, economic development, and the host nation's security forces. Although current Brigade level Unmanned Aerial Systems (UAS) can provide critical unadulterated views of progress with respect to these Logistical Lines of Operation (LLO), the majority of units continue to employ UASs for strictly conventional combat support missions. By incorporating these LLO targets into the mission planning cycle with a collective UAS effort, commanders can gain a decisive advantage in COIN. Based on the type of LLO, some of these targets might require more than a single observation to provide the maximum benefit. This thesis explores an integer programming and metaheuristic approach to solve the Collective UAS Planning Problem (CUPP). The solution to this problem provides optimal plans for multiple sortie routes for heterogeneous UAS assets that collectively visit these diverse secondary LLO targets while in transition to or from primary mission targets. By exploiting the modularity of the Raven UAS asset, we observe clear advantages, with respect to the total number of targets observed and the total mission time, from an exchange of Raven UASs and from collective sharing of targets between adjacent units. Comparing with the status quo of decentralized operations, we show that the results of this new concept demonstrate significant improvements in target coverage. Furthermore, the use of metaheuristics with a Repeated Local Search algorithm facilitates the fast generation of solutions, each within 1.72% of optimality for problems with up to 5 UASs and 25 nodes. By adopting this new paradigm of collective Raven UAS operations and LLO integration, Brigade level commanders can maximize the use of organic UAS assets to address the complex information requirements characteristic of COIN. Future work for the CUPP to reflect a more realistic model could include the effects of random service times and high priority pop-up targets during mission execution.
Thesis (S.M. in Transportation)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, 2012.Cataloged from PDF version of thesis.Includes bibliographical references (p. 129-132).
DepartmentMassachusetts Institute of Technology. Dept. of Civil and Environmental Engineering.
Massachusetts Institute of Technology
Civil and Environmental Engineering.