Proceedings of the 82nd Military Operations Research Society (MORS) Symposium, Chemical, Biological, Radiological, Nuclear and Advanced Explosives (CBRNE) Defense working group, Alexandria, VA (June 2014)

Unmanned robotic systems are a vital component of today’s military operations. They are used for operational maneuver (e.g., reconnaissance and IED defeat), maneuver support (e.g., route clearance and CBRNE), and sustainment (e.g., explosive ordnance disposal (EOD), convoy, and resupply) missions. As robots are used for more dangerous tactical operations that pose significant threats to soldiers, it is vital to increase their effective control range on the battlefield. Longer distance control and data feed downlink capabilities have become a tactical necessity in environments where tethers are not feasible and required standoff distances or complex terrain features limit line of sight communications. The construction of a wireless mesh network using additional UGVs with autonomous navigational capabilities provides a solution. In this paper, we present our approach to extending tactical range for IED and bomb disposal using autonomous ground robots. We combine computer vision, dismount detection and tracking, and tactical systems control to autonomously maneuver a relay robot to extend the effective stand-off range of IED and bomb disposal robots in complex terrain. The relay robot can also provide a dynamic second angle viewing perspective of delicate robot arm manipulations when cameras onboard the primary robot prove insufficient. The relay robot proactively tracks and follows a tele-operated UGV (lead robot) while reactively and proactively positioning itself to optimize relay communications between the lead robot and its control station. Our approach involves first generating a 3D map of the operational area based on inputs such as GPS positions, radio signal strength, and signal quality measures. We then predict the lead robot’s future path based on its known mission objectives and previous track data. Next, the relay robot’s path is optimized for its future ability to effectively relay signals back to the control station. Finally, we execute maneuvers using the open source Robotic Operating System to reposition the relay robot while routing the relay robot around obstacles. Our approach drastically increases the effective range of the lead robot, but also has the ability to provide a dynamic second angle perspective on manipulator arm actions and provides the operator with an up-close visual picture of the lead robot and its surroundings while it is beyond line of sight from the operator.