Phantom Space taps Ubotica for edge AI in space
Ubotica in Ireland is to supply edge AI technology for a satellites being developed by Phantom Space in the US.
The deal will allow Phantom Space to provide near-real time analysis of images from Earth Observation satellites without wating for processing on the ground.
Ubotica already supplies its CogniSAT-XE2 AI accelerator and CogniSAT-CRC compression system on nine active space missions in orbit.
The CogniSAT-XE2 On-Board AI Payload Coprocessor is a PC/104 form-factor for SmallSat and CubeSat missions built around the Intel Movidius Myriad X CV and AI COTS Vision Processing Unit (VPU). This has 16 vector cores and application-specific hard blocks provide high-performance parallel and hardware accelerated compute within a 1W power envelope.
Common Neural Network (NN) frameworks such as TensorFlow, PyTorch and Caffe can be used for NN model development and training, with the model subsequently compiled to target the Myriad 2 device using Intel’s OpenVINO toolkit.
Workflows contained within the CogniSAT-TK package guide this process to reduce error and the CogniSAT-XE2 uses pre-qualified models and layers available within OpenVINO. Pre-trained OpenVINO models can be used with transfer learning, or can be deployed directly on the CogniSAT-XE2 board using the Host Control Software (HCS) which is supplied as part of the package.
A host loads NN models to the CogniSAT-XE2 board, subsequently sending image frames and receiving processing results. The host can dynamically update the CogniSAT application over the primary interface, enabling flexible runtime solutions.
Ubotica supplies a range of utilities as part of the CogniSAT-TK package to operate seamlessly with most sensors currently on the market. Ubotica and deployment to the board involves the transfer of only a single configuration file, and runtime updates enable the updating of pipelines without requiring re-compiles or system reboots.
Adding the AI to the spacecraft dramatically reduces data bottlenecks, cuts transmission costs, and enable high-speed decision-making from orbit.
“We’re bringing intelligence to the edge, literally,” said Fintan Buckley, CEO and Founder of Ubotica. “Our AI meets Phantom’s platform right where the action happens, unlocking a new class of agile, responsive satellites.”
“Ubotica’s cutting-edge tech is a perfect fit for Phantom Cloud. Together, we’re creating a space ecosystem that thinks faster, costs less, and delivers the kind of real-time insight that makes a difference on the ground,” said Jim Cantrell, CEO and Co-founder of Phantom Space.
The partnership also expands access to third-party satellite operators, offering plug-and-play onboard processing even for missions lacking native compute power. This can be used for detecting wildfires, tracking vessels at sea, or supporting global communications.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
