Integrated Air/Ground Vehicle System for Semi-Autonomous Off-Road Navigation

Abstract

Current unmanned vehicle systems enable exploration of and travel through remote areas, but demand significant communications resources and constant human operation. DARPA and the US Army have recognized these limitations and are now pursuing semi-autonomous vehicle systems in the Future Combat Systems (FCS) program. FCS places high demands on robotic systems, which must assess mobility hazards under all weather conditions, day and night, in the presence of smoke and other airborne obscurants, and with the possibility of communications dropouts. Perhaps the most challenging mobility hazard is the "negative obstacle", such as a hole or ditch. These hazards are difficult to see from the ground, limiting maximum vehicle speed. From the air, however, these obstacles are often far easier to detect. In this context, we present a novel semi-autonomous unmanned ground vehicle (UGV) that includes a dedicated unmanned air vehicle - a "Flying Eye" (FE) - that flies ahead of the UGV to detect holes and other hazards before the onboard UGV sensors would otherwise be able detect them. This concept is shown in Figure 1. The FE can be considered a longer-range ?cout?to explore terrain before the UGV must traverse it, allowing the UGV to avoid certain areas entirely because of hazards or cul-de-sacs detected by the FE. The UGV itself carries its own extensive sensor suite to address the broad range of operating conditions at slower speeds and to confirm the hazards seen from the air. In the fully developed system, the UGV will deploy the FE from a landing bay on the back of the UGV. For covert operations, the FE will return to its landing bay on the back of the UGV. This paper presents the current prototype system as well as initial field experiments on the performance of the system.