Search for: All records

In recent years, networked airborne computing (NAC) has emerged as a promising paradigm because it can leverage the collaborative capabilities of unmanned aerial vehicles (UAVs) for distributed computing tasks. Despite the burgeoning interests in NAC and UAV-based computing, many existing studies depend on over-simplified simulations for performance evaluation. This reliance has led to a gap in our understanding of NAC’s true potential and challenges. To fill this gap, this paper presents a comprehensive approach: the creation of a realistic simulator and a novel hardware testbed. The simulator, developed using ROS and Gazebo, emulates networked UAVs, focusing on resource-sharing and distributed computing capabilities. This tool offers a cost-effective, scalable, and adaptable environment, making it ideal for preliminary investigations across a myriad of real-world scenarios. In parallel, our hardware testbed comprises multiple quadrotors, each equipped with a Pixhawk control unit, a Raspberry Pi computing module, a real-time kinematic (RTK) positioning system, and multiple communication units. Through extensive simulations and hardware tests, we delve into the key determinants of NAC performance, such as computation task size, number of UAVs, communication quality, and UAV mobility. Our findings not only underscore the inherent challenges in optimizing NAC performance but also provide pivotal insights for future enhancements. These insights encompass refining the simulator, reducing computation overheads, and equipping the hardware testbed with cutting-edge communication devices.

 

This paper considers the single source shortest path (SSSP) problem, which is the key for many applications such as navigation, mapping, routing, and social networking. Existing SSSP algorithms are designed mostly for shared-memory systems. Nevertheless, with the prevalence of diverse smart devices like drones, there is a growing interest in deploying SSSP algorithms over distributed computing systems so that they can run efficiently onboard of smart devices via Mobile Ad Hoc Computing or at the network edges via Mobile Edge Computing. In this paper, we introduce a communication-efficient ∆-stepping algorithm for distributed computing systems. The proposed algorithm is featured by 1) a message coordination architecture for reducing message exchanges between workers, 2) a pruning technique for reducing redundant computations, and 3) an aggregation technique for further reducing message exchanges when communication delay is significant. Theoretical analyses and experimental studies on real-world graph datasets demonstrate the promising performance of proposed algorithm. 

The integration of onboard computing capabilities with unmanned aerial vehicles (UAV) has gained significant attention in recent years as part of mobile computing paradigms such as mobile edge computing (MEC), fog computing, and mobile cloud computing. To enhance the performance of airborne computing, networked airborne computing (NAC) aims to interconnect UAVs through direct flight-to-flight links, with UAVs sharing resources with each other. However, despite the growing interest in NAC and UAV-based computing, existing studies rely heavily on numerical simulations for performance evaluation and lack realistic simulators and hardware testbeds. To fill this gap, this paper presents the development of two NAC platforms: a realistic simulator based on ROS and Gazebo, and a hardware testbed with multiple UAVs communicating and sharing computing resources. Through simulation and real flight tests with two computation applications, we evaluate the platforms and examine the impact of mobility on NAC performance. Our findings offer valuable insights into NAC and provide guidance for future advancements.