Asymmetric multiprocessing techniques in smart devices: Application in a drone navigation system

Smart devices can be considered an essential element of the Internet-of-Things (IoT) revolution. Devices able to gather and process data to provide high-level information to other nodes will enable advanced and effective smart applications in the near future. To let IoT objects to become smarter and smarter, advanced embedded architecture must be considered. In this direction, this paper exploits the use of heterogeneous Asymmetric MultiProcessing (AMP) techniques in the IoT scenario by considering a smart device in which real-time acquisition and processing capabilities are necessary: a drone. In the paper an AMP-based drone system is first presented, considering a System-on-Chip (SoC) solution based on microprocessors and microcontrollers. Then time performance in navigation parameters acquisition and processing is presented, considering a real-time operating system running on the core delegated to such task. Presented results show the benefits of the proposed approach in which high real-time constraints are satisfied in calculating navigation parameters, while other demanding IoT tasks can be executed simultaneously. Along with the selected use case, the proposed approach can be used in a wide range of smart devices in which higher computational performance are required.