Co-Design of Networking and Decentralized Control to Enable Aerial Networks in an Uncertain Airspace
pdf
Abstract:
Airborne networking utilizes direct flight-to-to-flight communication for flexible information sharing, safe maneuvering, and coordination of time-critical missions. It is challenging because of the high mobility, stringent safety requirements, and uncertain airspace environment. This project uses a co-design approach that exploits the mutual benefits of networking and decentralized mobility control in an uncertain heterogeneous environment. The approach departs from the usual perspective that views physical mobility as communication constraints, communication as constraints for decentralized mobility control, and uncertain environment as constraints for both. Instead, we proactively exploits the constraints, uncertainty, and new structures with information to enable high-performance designs. The features of the co-design such as scalability, fast response, trackability, and robustness to uncertainty advance the core CPS science on decision-making for large-scale networks under uncertainty. Three directions are pursued under this framework. First is a random mobility model framework that bridges networking and decentralized mobility control under uncertainty. Second is an uncertainty-exploiting decentralized mobility control to facilitate robust networking. Third is a practical networking scheme to facilitate fast decentralized mobility control under uncertainty. In the first few months of the project, we made substantial effort in recruiting student and staff for this position. Some of the positions will be filled from January 2016. We also explored the capability of M-PCM-OFFD (a method we developed to evaluate high dimensional uncertainty) in facilitating optimal control under uncertainties. Instead of evaluating over the whole range of uncertainties, we only need to evaluate a few samples within this range, so as to significantly reduce the computational costs. Such cost reduction is to achieve real-time control of large-scale systems. We conducted theoretical analysis and proved that under three broad cases, the M-PCM-OFFD-based optimal control achieves the optimal solution of those standard stochastic optimal control methods but with much reduced computation cost.
Submitted by Yan Wan
on
Abstract:
Airborne networking utilizes direct flight-to-to-flight communication for flexible information sharing, safe maneuvering, and coordination of time-critical missions. It is challenging because of the high mobility, stringent safety requirements, and uncertain airspace environment. This project uses a co-design approach that exploits the mutual benefits of networking and decentralized mobility control in an uncertain heterogeneous environment. The approach departs from the usual perspective that views physical mobility as communication constraints, communication as constraints for decentralized mobility control, and uncertain environment as constraints for both. Instead, we proactively exploits the constraints, uncertainty, and new structures with information to enable high-performance designs. The features of the co-design such as scalability, fast response, trackability, and robustness to uncertainty advance the core CPS science on decision-making for large-scale networks under uncertainty. Three directions are pursued under this framework. First is a random mobility model framework that bridges networking and decentralized mobility control under uncertainty. Second is an uncertainty-exploiting decentralized mobility control to facilitate robust networking. Third is a practical networking scheme to facilitate fast decentralized mobility control under uncertainty. In the first few months of the project, we made substantial effort in recruiting student and staff for this position. Some of the positions will be filled from January 2016. We also explored the capability of M-PCM-OFFD (a method we developed to evaluate high dimensional uncertainty) in facilitating optimal control under uncertainties. Instead of evaluating over the whole range of uncertainties, we only need to evaluate a few samples within this range, so as to significantly reduce the computational costs. Such cost reduction is to achieve real-time control of large-scale systems. We conducted theoretical analysis and proved that under three broad cases, the M-PCM-OFFD-based optimal control achieves the optimal solution of those standard stochastic optimal control methods but with much reduced computation cost.