Hamiltonian dynamics and control of a joint autonomous land–air operation

Abstract

Rapid technological advances in artificial intelligence and additive manufacturing have fueled the speculation that large numbers of cheap autonomous robots swarming to achieve mission objectives with minimal human control will revolutionize search-and-rescue, disaster relief and humanitarian assistance operations. These kinds of operations usually occur in a GPS-denied and communication austere environment, where only nearest-neighbors networking among the autonomous robots may be possible. Hence, the ability to predict and control the real-time global dynamic behavior of this high-dimensional, highly complex and highly nonlinear, decentralized autonomous system, immersed in an uncertain and changing environment—and yet successfully performing at the edge of chaos—will be critical. This paper proposes a rigorous computational framework for prediction and control of a large-scale joint swarm of unmanned ground vehicles (UGVs) and unmanned aerial vehicles (UAVs), performing a joint autonomous land–air operation. We present two cases for this joint UGV + UAV swarm: (1) a discrete Hamiltonian dynamics and affine control over Euclidean groups of individual UGVs and UAVs in a swarm of hundreds of vehicles and (2) a continuous 2-soliton ‘signature’ evolution of a joint UGV and UAV swarm of thousands of vehicles.