Every year, natural disasters disrupt millions of lives in India. Floods, glacial bursts and landslides often strike without warning, leaving officials scrambling to organise relief. The damage is measured not just in lives lost but also in homes destroyed and livelihoods interrupted.
One of the biggest challenges is speed. Emergency teams often wait hours—or even days—for reliable ground information before they can act. By the time data arrives, the chance to prevent the worst damage has already slipped away.
Space-based observation has long promised a solution. Satellites can capture images of unfolding events across vast areas that would otherwise be hard to reach. Yet, the usefulness of those images is determined by how quickly they can be delivered.
GS Srinivas Reddy, former director of the Karnataka State Natural Disaster Monitoring Centre (KSNDMC), explained this to AIM with a concrete example: “Some satellites have a repetitive coverage of 15 days. If the same-day coverage happens, I can get the data the next day. Otherwise, I may have to wait two or three days.”
Often, satellite data crawls through a long pipeline—from orbit to ground stations and then through multiple layers of processing before it becomes usable. This delay is precisely the gap a Bengaluru space-tech startup is now trying to close.
SkyServe is working on onboard processing for satellites, technology that could shorten the time between capturing an image and turning it into actionable insights on the ground. Vishesh Vatsal, co-founder and CTO at SkyServe, said that this shift could make a world of difference in disaster relief.
“We’re talking about turning hours into minutes,” he said in an exclusive interview with AIM, pointing out how slow data transfers can delay action during floods, glacial bursts or infrastructure failures.
“Especially for disaster-related response, if we can get satellite data in minutes instead of hours, it will be very useful,” Reddy added.
Building Satellites That ‘Think’
Having studied aerospace engineering at IIT Kanpur and later worked on lunar descent systems at Team Indus, Vatsal said the idea for SkyServe came after he noticed how heavily government contracts dominated the space exploration market.
“It made sense to focus on commercial Earth observation.” One technology that stood out to the team was onboard processing. With deep learning advancing so quickly, Vatsal explained, it was only a matter of time before it broke through in space as well.
SkyServe’s core product is its onboard software platform called Storm, which works directly on the small computers carried by satellites. “We have the ability to receive data in a variety of pre-processed or semi-processed forms, and the in-house knowledge to make it to a more processed product that becomes one of our key value propositions,” he added.
By compressing and filtering out useless data, such as cloud-covered images, the satellite sends down only the information that truly matters. In effect, satellites stop being passive cameras in orbit and instead “think”, making real-time decisions about what data to retain, what to discard and when to alert users.
Vatsal explained that there would be a radical change in response times. If a disaster were to strike and data wasn’t immediately available, usable information could still be delivered by the end of the day, significantly reducing turnaround times.
SkyServe has already flown Storm on two missions and is preparing for a third. Along the way, the team has learnt to handle different sensor types and to overcome tough constraints around power and memory.
Smaller Packets of Data
Vishesh described his own motivation. “Psychologically, I tend to gravitate towards hard problems, and I enjoy [tackling] those.” His earlier work on vision-based navigation for lunar landings, he added, proved helpful in preparing for building real-time computer vision for satellites.
Reddy clarified that disaster monitoring in Karnataka has, since 2009, already relied heavily on real-time sensor networks, including rainfall gauges and river monitoring systems, down to the gram panchayat level.
He emphasised the need for satellite data alongside these sensors by saying, “Satellite data can be utilised in all aspects…[It can play a role] before disasters for monitoring and early warning, during disasters for rescue and relief, and after disasters to assess damages and [guide] reconstruction.”
In practice, SkyServe’s software compresses large image files in orbit. Vatsal explained that, for example, if a satellite were flying over Bengaluru on a cloudy day, most of the image might be cloud-covered. The software could strip away the clouds, segment the images and then check the remaining areas for signs of flooding.
By doing this, the company aims to create smaller areas of interest, which could help segregate the research into focus sectors. Reducing gigabytes of raw imagery to megabytes of processed results means satellites can send down valid data in a single pass instead of several.
Involvement with NASA and More
SkyServe has demonstrated its technology with Italy-based aerospace company D-Orbit and collaborated with scientists at NASA’s Jet Propulsion Laboratory (JPL). From the time of contract signing to actual execution in space, the turnaround took just three weeks, the company said.
SkyServe collaborated with Steve Chien, senior research scientist at JPL, to identify nine different models and standardise them on SkyServe’s search platform, demonstrating their capabilities. The goal was to explore satellite autonomy and climate-related event monitoring, with the potential for future missions together.
The company has also worked with IN-SPACe, the Indian government’s space regulator. Disaster management, he stressed, is one of the most urgent areas. “It’s almost like some things that can be solved are not being solved, especially for a country like India. I think the disasters, the value of loss that comes is slightly perceived to be low, but it’s something that’s very sad.”
What Comes Next?
SkyServe sees opportunities in agriculture, defence and urban planning, but disaster management remains the focus. Vishesh believes that multiple constellations working together could provide warnings days in advance.
Looking ahead, SkyServe aims to run its platform on higher resolution satellites. This will open up a lot of use cases for the company that would otherwise not be demonstrable. Reddy mentioned that startups must work with organisations like ISRO, IMD, National Disaster Management Authority and National Remote Sensing Centre to ensure their solutions are adopted. Else, he mentioned, negotiating for data after a disaster is very difficult. “It should be before disasters, in advance.”
The company sees progress as inevitable, with demand for real-time space data on the rise. As disasters intensify, SkyServe bets that satellites which ‘think’ for themselves can provide the speed that response teams need.
The post Satellites That ‘Think’ Could Change How India Responds to Disasters appeared first on Analytics India Magazine.
