AI to boost Earth observations onboard cubesats

April 27, 2020 // By Julien Happich
cubesats
During an online press conference held by Teledyne e2v to promote the company’s latest product developments, the group’s CMOS sensor division came into focus as being part of a new cubesat project launched by Grenoble’s University Space Centre (Centre Spatial Universitaire de Grenoble or CSUG).

First, application engineer Thomas Porchez presented the on-board processing solutions under development for such applications. With the 2-in-1 Qormino processing module soon to be launched, the company is integrating its radiation tolerant LS1046 processor with 4GB of radiation tolerant (RadTol) DDR4 memory on a single 44x26mm substrate with a 2.1GT/s built-in DDR4 bus (featuring ECC).

The Qormino processing platform developed by Teledyne e2v.

Built around a quad-core ARM Cortex A72 running at up to 1.6GHz, the Qormino offers 30 000 DMIPS of computing capability, Porchez highlighted, noting that a development kit was already available while the standalone LS1046 processor and DDR4 memory module were being tested for space grade reliability before the Qormino module would undergo such tests by the end of 2020 or early 2021.

Roadmap for the Qormino platform.

Next Eric Chevallier, Bid Manager at Teledyne e2v discussed the evolution of image sensors for space applications, acknowledging a strong shift from CCD devices (declining) to CMOS devices. The market shift leverages the availability of low-cost mass-produced Commercial Off-The-Shelf (COTS) CMOS sensors (including those from Teledyne but also some from third party manufacturers) up-screened and tested according to space grade standards.

Teledyne e2v’s space grade Emerald CMOS sensor.

Among Teledyne e2v’s image sensors getting traction for space applications, Chevallier highlighted the Emerald CMOS sensor family, with parts ranging from 8Mpixels at a 4Kx2K resolution with pixels 2.8µm per side to a 67Mpixel device delivering 8K UHD resolution (with 2.5µm pixels) and capable processing 60 frames per second at full resolution.

In the scope of the new cubesat project dubbed “QlevEr-Sat”, it is a 16Mpixel 4Kx4K (4096 x 4096) device that has been selected (with 2.8μm pixels) which can also be used as a 2kx2k matrix. The image sensor is available in B&W and color versions with 8, 10, or 12 bit resolutions and has a throughput of 47fps at 10 bit or 30fps at 12 bit.

The partnership between the CSUG and Teledyne e2v is not new, the two groups started their collaboration in 2017 with the ‘AMICal-Sat’ project, a 2U nanosatellite whose mission is to observe auroras and perform image captures. The imaging payload of this satellite is based on Teledyne’s Onyx 1.3MP CMOS sensor. While the nanosatellite is now ready for launch into orbite with the ArianeSpace VEGA launcher, its launch initially scheduled for the 24th of March 2020 has been delayed due to the coronavirus pandemic.

Started in the first quarter of 2020, the QlevEr-Sat collaborative project will leverage Teledyne e2V’s 16MP Emerald CMOS sensor together with the radiation tolerant Qormino processing module.

The QlevEr Sat aims to bring AI onboard a cubesat for
Earth observation.

Mathieu Barthelemy, Director of the CSUG, presented QlevEr-Sat as an Artificial Intelligence (AI) Earth observation demonstration satellite designed to test AI routines on board and drastically reduce ground data transmission.

To illustrate the benefits that AI could bring in image analysis and data reduction, Barthelemy noted that a satellite taking one 16Mpixel image every second (coded in 12 bits), ground data transmission amounted to 16 Tbits/day (2.03 TB/day). With AI and on board processing, the QlevEr-Sat could make autonomous decisions. With the Emerald CMOS imager selected for this project, the ultra small 2.8μm pixels provide a ground based resolution of 4.8m at 500km. For a 16Mpixels sensor, it means the satellite’s camera covers an 18 kilometers-wide swath of land below its orbital path.

The Qormino module onboard the QlevEr-Sat will run machine learning algorithms developed by Grenoble’s Multidisciplinary Institute in Artificial intelligence (MIAI) and trained for different applications.

Due to be launched circa Q1 or Q2 of 2022, the QlevEr-Sat is expected to operate for a year or two, surveying a specific region for damages from natural disaster such as volcanoes, earthquakes or tsunamis. The cubesat could carry out forest surveys on wild fires and deforestation or it could monitor urban expansion or other human activities associated with societal challenges.

Teledyne - www.teledyne.com

CSUG - www.csug.fr

 


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.