Optimal UAS Assignments and Trajectories for Persistent Surveillance and Data Collection from a Wireless Sensor Network

Developing a methodology for multiple unmanned aircraft assigned to fly optimal trajectories in order to survey and collect a pre-specified amount of data from a fixed, ground-based wireless sensor network.

October 1, 2022

Air Force Institute of Technology, Wright-Patterson Air Force Base, Ohio

DoD groupings of UAS by size and performance

The Department of Defense (DoD) estimates manpower cost is the largest component in the operation of Unmanned Aircraft Systems (UAS). From planning, controlling, supervising, analyzing, replanning, delivering data, and other functions, the human operator currently bares the majority of the burden for these tasks.

The most critical phases of mission profiles are often either manually performed or pre-programmed by human operators. These functions “include critical flight operations, navigation, takeoff and landing of unmanned aircraft, and recognition of lost communications requiring implementation of return-to-base procedures.”. Furthermore, UAS that conduct Intelligence, Surveillance, and Reconnaissance (ISR) missions often collect and deliver raw data. For example, live streaming video from a UAS requires human interpretation and analysis before it can be used for decision making.

Ideally, future unmanned systems will be capable of many of these tasks autonomously, moving from pre-programmed to situational, decision-based maneuvers and from simply delivering raw data to directly providing actionable intelligence. The DoD also predicts, “autonomous systems may even optimize behavior in a goal-directed manner in unforeseen situations (i.e., in a given situation, the autonomous system finds the optimal solution).” However, the UAS itself is limited by its on-board capabilities. Networking the UAS with standoff components, such as other UAS, manned vehicles, and remote sensing create a force multiplying effect, dramatically improving a single UAS’s capabilities.

To meet the DoD’s vision for future UAS capabilities and lower manpower costs, systems incorporating UAS and standoff components are needed. Rapidly maturing hardware require novel algorithms to enable UAS to autonomously collect and analyze data, make decisions based on collected data, and deliver pertinent information to the end-users when their data are efficiently collected. Advanced wireless sensor networks (WSN) can provide time-critical and precise localized environmental information critical to decision making. Researching how to smartly combine multiple UAS, WSN, and algorithmic components motivates this research.

The concepts of wireless sensor networks, combined with efficient data collection, and network energy management collectively compose a relatively new field of study. The term “Wireless Sensor Network” began to appear in literature in the late 1990’s, specifically in Pottie in 1998. Immediately, the inherent problems of data collection and network energy management were apparent and became a focus of this burgeoning research area. Early work focused on node-to-node multi-hopping data routing protocols within the network. However, as WSNs grew and matured, new applications began to appear such as forest fire detection and monitoring, environmental mapping, traffic monitoring and tracking, and in the military, battlefield surveillance. These new applications further complicated data collection and network energy management as well as introducing newer challenges like network deployment, size, cost, upgrades, usable lifespan, survivability, and adaptability to changing network requirements.

Alternative data collection methods reduced network energy consumption rates and collected more data by using robots, a.k.a. mobile collection agents, to physically visit and survey sensor nodes within the network, collect their data, and deliver or relay the data back to the user or data sink. The mobile agents were assumed to have larger storage capability, which in some architectures freed the WSN of the multi-hopping protocol, often the largest consumer of network energy reserves.

This work was performed by Nidal M. Jodeh for the Air Force Institute of Technology. For more information, download the Technical Support Package (free white paper). AFIT-0009