What do system-on-chip (SoC) based flight controllers do?

Safely flying UAVs beyond visual line of sight (BVLOS) requires the UAV to not only be cognitively aware of its surroundings, but also intelligent and intuitive in choosing real-time flight courses.

Determining how to handle unexpected situations—such as navigating among dense buildings and other man-made structures—entails a great deal of processing. Today, many industrial-use drones fly at 85 mph or more; for these drones, most processing is done on board and quickly, since there may be lack of time for the vehicle to generate messages and receive instructions about an unexpected situation.

To make UAVs truly cognitive, an array of sensors must be embedded throughout the vehicle. These sensors generate a high volume of data, which requires substantial computing capabilities and real-time processing ability. We believe that System-on-Chip (SoC) are ideally suited to perform these tasks. Field programmable gate arrays (FPGA) meets the strict timing requirement of real-time signal processing, while the dual-core ARM processor handles the complicated algorithm designed for specialized applications. This control solution functions as the brain of the smart drone, providing clear advantages over the microcontroller unit (MCU) in terms of processing power and I/O capability.

Why are SoC controllers better?

AI-Capable Processing Power:
With the drone industry quickly reaching new commercial and consumer markets with creative and advanced applications such as precision agriculture, 3D modeling/mapping/surveying, delivery services, industrial inspection, and more, drone applications are becoming increasingly complex and require greater processing power. Ainstein’s SoC controllers support complex applications with load-balancing, FPGA logic which enables sensor fusion, real-time data processing and deep learning; these functions free up the CPU to better perform high-level autonomous decision making
and flight control.

Unmatched Flexibility:
The SoC controllers of our sister company for drone products, Aerotenna, offer one of the most flexible computing platforms, and can process radar, vision, and other sensing inputs; they are equipped with over 30 programmable I/Os which support most standard interfaces. Built on the configurable nature of FPGAs, Aerotenna OcPoC flight controllers can be reprogrammed in order to support the latest algorithms, and are also able to be customized for various performance requirements without complete custom hardware. They have ASIC capabilities, but with convenient flexibility for changes.

Industrial Grade Redundancy:
OcPoC flight controllers are designed to function in cases of unexpected system component failures during flight. To ensure continued flight for urgent missions, triple-redundancy for critical components—such as GPS, IMU, and more—are built into the system.

Developer Friendly:
The SoC controllers of our sister company for drone products, Aerotenna, can run a variety of embedded operating systems, such as Linux, making it easy to develop and modify. The controllers support popular open-source software suites, such as PX4 and ArduPilot, and come with proprietary collision-avoidance algorithms. We simplify the development process for developers, putting every idea within reach.

Can Ainstein’s technology be used for security applications?

Yes! Radar technology can be used to not only detect obstacles for drones, but also to detect moving people or vehicles in applications such as corrections management, perimeter control, and border control. As these applications are highly environment-dependent, please contact us directly to discuss your particular needs with us.