Ubotica's LMI platform, unveiled in April 2026, enables satellites to analyze imagery and make decisions in orbit, converting raw camera feeds into real-time maritime alerts . The system's key capabilities include:
Designed for dual-use (civil and defense) maritime security, the platform targets the following threats :
| Traditional EO Systems | Ubotica's Orbital AI |
|---|---|
| Satellites act as passive data collectors, capturing images and storing them for later downlink | AI inference runs directly on the satellite's hardware, processing imagery in real time on orbit |
| Raw imagery must be transmitted to a ground station and then processed by analysts — a process that can take hours to days | The satellite decides what is relevant, discards cloudy or irrelevant pixels, and only sends back actionable intelligence within minutes |
| Satellites cannot dynamically reprioritize observations; they follow a fixed tasking plan uploaded from the ground | Satellites can autonomously retask based on events detected in orbit, enabling a "sense-and-respond" capability |
| Bandwidth-limited: massive raw data volumes create bottlenecks in downlink | On-orbit processing drastically reduces data volume, cutting latency and downlink costs |
Ubotica's underlying technology, branded SPACE:AI, was first flight-proven in 2020 as the first hardware-accelerated AI inference of Earth observation images on an in-orbit satellite, and has since been refined through missions like CogniSAT-6 and partnerships with NASA and ESA .
The funding will be used to accelerate commercial deployment of LMI, expand sales to government and defense customers, and scale the rollout of Orbital AI capabilities across additional satellite constellations .
Ubotica's LMI platform, unveiled in April 2026, enables satellites to analyze imagery and make decisions in orbit, converting raw camera feeds into real-time maritime alerts . The system's key capabilities include:
Designed for dual-use (civil and defense) maritime security, the platform targets the following threats :
| Traditional EO Systems | Ubotica's Orbital AI |
|---|---|
| Satellites act as passive data collectors, capturing images and storing them for later downlink | AI inference runs directly on the satellite's hardware, processing imagery in real time on orbit |
| Raw imagery must be transmitted to a ground station and then processed by analysts — a process that can take hours to days | The satellite decides what is relevant, discards cloudy or irrelevant pixels, and only sends back actionable intelligence within minutes |
| Satellites cannot dynamically reprioritize observations; they follow a fixed tasking plan uploaded from the ground | Satellites can autonomously retask based on events detected in orbit, enabling a "sense-and-respond" capability |
| Bandwidth-limited: massive raw data volumes create bottlenecks in downlink | On-orbit processing drastically reduces data volume, cutting latency and downlink costs |
Ubotica's underlying technology, branded SPACE:AI, was first flight-proven in 2020 as the first hardware-accelerated AI inference of Earth observation images on an in-orbit satellite, and has since been refined through missions like CogniSAT-6 and partnerships with NASA and ESA .
The funding will be used to accelerate commercial deployment of LMI, expand sales to government and defense customers, and scale the rollout of Orbital AI capabilities across additional satellite constellations .