A new chapter in on-board data processing
Leopard is a CubeSat standard-compliant Data Processing Unit which enables mission designers to apply Artificial Intelligence solutions in space. It was designed to support capturing, managing and processing of data in orbit. Leopard redefines the current approach to remote sensing.
Now, instead of sending huge, unprocessed sets of data to ground stations, Leopard uses Deep Neural Networks to process data on-board and therefore only sends the most important and valuable insights to the ground. By reducing the time and cost of data transfer and processing, it enables you to focus on a rapid response to any detected phenomena.
Are you interested in on-board data processing?
Key advantages
During mission operations
On-board image and data processing including real-time classification, calibration and data compression
During mission development and testing
Uploading new algorithms while in orbit, as well as on-orbit updates and fixes
Multi-purpose mission
Detection of different events within one mission and reconfiguration of the same satellite for a new purpose
Leopard is integrated with a powerful FPGA to accelerate the execution of deep learning algorithms and has a throughput of up to 3 Tera Operations Per Second. A number of hardware and software measures protect the computer against the influence of radiation.
With its extraordinary capabilities-to-size ratio (less than 1U), wide supply voltage range and universal interfaces, it is compatible with most CubeSats platforms. Its scalable and customisable architecture makes it possible to create larger and more powerful versions dedicated to bigger platforms as well.
Leopard is a part of Smart Mission Ecosystem– hardware, software and AI-powered algorithms designed to complete your mission.
Case study
Leopard will be utilized by the Intuition-1 satellite, coupled with a 150-band hyperspectral sensor to perform image segmentation and object detection. For the purpose of on-board data processing during the mission, we will use two high-performance nodes for redundancy and parallel operation.
Leopard is perfectly compatible with The Herd – algorithms dedicated to Earth Observation which are responsible for on-board data processing and pre-processing.
Key applications
Image segmentation and object detection
Signal quality enhancement
Data compression and encryption
Spacecraft autonomy
Optical navigation
Many more!
Do you want to find out more about implementation of AI in your mission?
TECHNICAL SPECIFICATION
| PROCESSING CORES | Zynq UltraScale+ MPSoC: ZU6EG| ZU9EG| ZU15EG |
|---|---|
| Quad ARM Cortex-A53 CPU up to 1.5 GHz | |
| Dual ARM Cortex-R5 in lock-step | |
| FPGA for custom function implementation | |
| MEMORY | 4-16 GiB DDR4 providing with ECC |
| 4-16 GiB SLC flash-based file system storage (EDAC) | |
| Up to 2x256 GiB SLC flash-based data storage | |
| INTERFACES | Interfaces: CAN, LVDS, SPI, RS422/485, UART, GTY and GTH transceivers |
| Additional customisable interfaces upon request: SpaceWire, Ethernet | |
| LVDS/RS422 interfaces compatible with X/S-Band radios and CCSDS-compatible communication channel upon request | |
| SPECIFICATIONS | A radiation hardened Payload Controller |
| A radiation hardened Payload Controller | |
| Power Consumption 7.5 W to 40 W depending on workload and specified processing speed | |
| Computational Throughput for Neural Networks Processing up to 3 TOPS | |
| Thermal interface customisable for satellite architecture | |
| SOFTWARE ECOSYSTEM | 64-bit Linux |
| Deep Learning Accelerator fed with Caffe or TensorFlow models | |
| Fully reconfigurable in orbit | |
| REDUNDANCY | Possibility to introduce additional redundancy to each version |
| FORM-FACTOR | CubeSat standard-compatible, < 1U |