A Cooperative Autonomous System for Heterogeneous Unmanned Aerial and Ground Vehicles

Abstract

Today’s military intelligence, surveillance, and reconnaissance (ISR) missions routinely employ unmanned aerial vehicles (UAV) and unmanned ground vehicles (UGV) as versatile, real-time, remote sensing platforms. One of the main challenges of deploying such unmanned systems for large-scale military operations is the lack of interoperability among different types of unmanned systems. This constraint introduces even greater problems as we move toward unmanned missions that require multiple platforms to work cooperatively, even in environments where human operators cannot readily provide any mission inputs. If we expect cooperative unmanned systems technologies to advance rapidly in the future, it is essential that we, the unmanned systems community, remove this interoperability constraint. To this end, we developed a scalable, onboard unmanned system called cooperative autonomous system (CAS) capable of mounting on multiple vehicle types to address key aspects of the interoperability problem. The CAS includes a comprehensive collection of hardware and software modules designed for real-time coordination of unmanned aircraft and autonomous ground vehicles. These modules are essential capabilities to perform cooperative missions: to navigate and control the platform, to communicate with other unmanned systems, to process information transmitted by neighboring unmanned systems, and to fuse data collected from other unmanned systems with its local sensor data. We created CAS interfaces for several commercially available autopilots, mobile robot platforms, and industry-standard sensor and communication devices. The CAS software is based on an open-source, multi-threaded, event-driven framework that combines cross-platform compatibility with object-oriented software that encapsulates a variety of asynchronous control, sensing, and communication mechanisms. The hardware of CAS is made of modular units to allow a plug-and-play ability, depending on a particular mission application. We successfully demonstrated the interoperability of CAS by implementing the system on two different unmanned aircraft and an unmanned mobile ground platform. We show results from an ISR experiment with three different mobile platforms and a set of stationary sensors. Sensors employed were electro-optical and infrared cameras as well as radio frequency receivers.