SkyServe tests AI models for disaster monitoring in space

SkyServe, an Indian startup that focuses on edge computing, is collaborating with NASA’s Jet Propulsion Laboratory (JPL) to test artificial intelligence models. These tests are being conducted on a satellite managed by D-Orbit. The AI models are part of a project called New Observations Strategies Sensorweb, aimed at monitoring environmental issues like wildfires, floods, and urban heat. In March, SkyServe completed its tests using JPL’s AI models on its STORM edge-computing software. This software was installed on a D-Orbit ION Satellite Carrier located in low-Earth orbit. Vinay Simha, the founder and CEO of SkyServe, stated that the mission involved uplinking the software from a ground station and running the AI models on the satellite. He described this as a new method for observing Earth. The AI models focused on improving the way water and vegetation data are processed to make them easier to download. The successful operation of these models relies on advanced edge computing technology, which helps integrate and optimize AI across various sensor setups. SkyServe's technology simplifies the use of AI models across different satellite platforms, according to Vishesh Vatsal, the company’s chief technology officer. This ensures that the deployment of AI is consistent and efficient. SkyServe’s platform connects advanced AI models with the varied hardware of satellite systems for better Earth observation. D-Orbit has plans to equip its ION satellite carriers with SkyServe's STORM technology. Similarly, Loft Orbital has utilized SkyServe's edge computer on its spacecraft. Additionally, another STORM system has been integrated into a cubesat launched in December through India’s Polar Satellite Launch Vehicle. SkyServe also offers Surge, a platform for developing and testing AI models on the ground.