A brain-inspired approach for resilient AI processing
Texas A&M University
Researchers in the Department of Electrical and Computer Engineering at Texas A&M University have received a two-year, $1.2 million grant from the U.S. Army Research Laboratory (ARL) to explore a new approach to cloud computing in battlefield environments.
Led by Drs. I-Hong Hou, Krishna Narayanan, P.R. Kumar and Dileep Kalathil, the project aims to revolutionize a growing challenge in modern computing: how to deliver the power of artificial intelligence (AI) not just from distant cloud servers, but directly to users and devices operating in constrained, dynamic, or infrastructure-poor environments.
Cloud-based AI tools like ChatGPT are common in civilian life. A user types or speaks into their phone, which sends input across the internet to a massive data center, often located far away. There, the AI model runs on powerful computers, processes the request, and sends back a response.
While this architecture works well for many everyday applications, it reveals serious limitations when connectivity is constrained, latency is critical, or robustness is paramount.
“In remote or time-sensitive scenarios, especially in battlefield applications, that kind of setup can be a real problem,” Hou said. “Even just a few seconds of delay could mean the difference between success and failure or even life and death. Whether it’s a military mission, disaster response, or autonomous system, the need is the same: intelligent decision-making must happen in real-time, close to the data.”
Rather than relying on one central data center, every device — Wi-Fi access points, routers, cellular towers, even vehicles — will play a role in both moving and processing data. This eliminates a single point of failure, making this method more secure and reliable.
"It's similar to how our brains work," Hou said. "Each brain cell, or neuron, does a little bit of thinking and also passes signals along to other neurons. That’s what we want to build, a distributed system where every part of the network helps with both communication and computation."
Through this project, every network component becomes a tiny computer. These components can handle small bits of processing independently, passing partially processed data to the next device. By the time the data reaches a more powerful computer, like a server in a vehicle or a base station, it’s already mostly processed, making the final computations fast and efficient.
The team’s new system offers speed and resilience. Since processing happens close to the action, the system can deliver results almost instantly. There is also no single weak spot. If one part of the network is damaged or removed, the system can potentially adapt and reroute tasks elsewhere.
"Every device does a little bit, and if one goes down, others pick up the slack," Hou said. "It makes the entire system much harder to break."
One major goal of this project is to design network processing, where data is partially processed step-by-step as it moves through the network. This requires coordinating a wide range of devices to work together intelligently and efficiently.
Another key focus is resilience and safety. Because the network supports safety-critical applications, it must be able to withstand attacks, adapt to changing missions, and continue operating under extreme conditions.
“We’re doing this because we want AI to make smart decisions in real-time,” Hou said.
While the project is designed to support ARL applications — where robust, real-time AI in contested environments is essential — the broader implications are far-reaching. The same principles apply to smart cities, autonomous vehicles, remote sensing, industrial robotics, and environmental monitoring.
By Katie Satterlee, Texas A&M University College of Engineering
###
British Army will use AI to strike enemy targets on battlefield by 2027
The £1bn Asgard system will use a network of sensors to detect enemy targets before employing AI to recommend action - within minutes or even seconds
The British Army will rely on AI to detect and strike enemy targets on the battlefield in just two years
The £1bn system, known as Asgard, uses a network of sensors to detect enemy targets or incoming attacks, then employs AI to identify them and recommend a course of action.
Traditionally, information on targets and threats is collected and inputted manually, which can take several hours.
Now that information will be gathered by AI, along with options for military action, and presented to officers within minutes or even seconds.
The project is due to be fully operational by 2027, when the head of the Army has warned the UK needs to be ready to fight or deter a war against Russia, potentially supported by China, Iran and North Korea.
The current Asgard system has a “human in the loop” to identify friendly vs enemy assets and make the final call on military strikes, but officials suggest that this could change in future.
The system is technically capable of running without human oversight and insiders did not rule out allowing the AI to operate independently if ethical and legal considerations changed.
However, at present there are no plans to remove humans from the system.
Asgard would be rolled out in case of a war or during military exercises.
Some of the system will be shared with allies to bolster cooperation, while others will be sold as an export to boost domestic economic growth.
Officials said the Army was learning from the war in Ukraine, where soldiers are already using AI and software to speed up the process of identifying and hitting Russian targets.
‘Every citizen can play their part’
One of the 27 firms contributing to Asgard is British-Estonian firm SensusQ, co-founded by serving Estonian reservist Villiko Nurmoja.
With a 183-mile border with Russia and a history of Soviet occupation, Estonia has seen firsthand the threat posed by Vladimir Putin.
It is also home to the largest contingent of British troops deployed anywhere in the world.
Nurmoja’s firm has created an app in which civilians and military alike can feed in information about suspicious objects.
“The app can be used even in peacetime; if you see a crime, you can report it. But in a military context, if you see the enemy aircraft coming, or enemy drones coming or an enemy tank or convoy is moving somewhere, you take a picture and send it. Every citizen can be a sensor, not only soldiers,” he said.
While the app might sound high-tech, Nurmoja describes it as a modern version of civilian reporting during the Second World War.
Thanks to his military service, Nurmoja understands the difficulties faced by troops defending Nato’s eastern border and is attempting to solve them for the next generation of soldiers.
“We started to build the intelligent management system that we missed during our service years; instead of using seven/eight different systems on three to four different computer systems at the same time, I want to have one thing,” he said.
“The amount of human hours you spend on putting intelligence together even on a lower tactical level, it’s enormous.”
Smartphone microphones can warn of drone attacks
Another firm bringing familiar technologies to the high-tech Asgard system is Mind Foundry, which uses smartphone microphones to detect and warn of drone attacks, which are causing around 70 per cent of casualties in Ukraine.
The firm specialises in using acoustic signals to detect threats. It has previously used smartphone microphones to predict whether mosquitoes were carrying malaria, according to the wingspan, location and time of year, and underwater signals to detect maritime threats.
The new system, which can be downloaded as an app, buys soldiers crucial seconds to prepare for incoming drone attacks and helps the military gather intelligence on the nature of the enemy’s drone operations.
“Every soldier will be carrying something called an attack device, which is like a mobile phone sitting on their chest giving them streams of information. We deploy onto that, so a soldier would see an alert telling them that our software detected a drone coming towards them,” said Al Bowman, former director of the Army’s intelligence centre and now general manager of Mind Foundry.
“We can feed that back to a central control point, so you’d be able to see what every soldier’s device is listening to, so you can then use that to track everything and determine what’s coming next.”
The app monitors the sound of the drone wings to identify the type of drone, direction of travel and proximity up to 50m away.
“If you can detect what drones are coming, then you can start to understand what the target is and do something about it, and eventually counter that by understanding how the adversary is using drones,” Bowman said.
‘AI can make eight-hour process take 23 seconds’
Many of the contractors involved in Project Asgard use AI to save time and boost battlefield efficiency.
Shefali Sharma, director of Oxford Dynamics, said her AI-driven system – which collates and organises information and recommends different courses of action – can complete a process which would usually take an analyst eight hours in just 23 seconds.
“The particular problem that we were set was to look at open-source intelligence as well as the documents or information that analysts have available on their systems. How can we pull all that data together to provide tactical information to those operators and users in the shortest amount of time possible?”
The firm is not planning to eradicate the human in the loop, but to use AI to allow soldiers to focus on things that they can do uniquely well.
“We don’t believe that these systems should be allowed to make decisions autonomously. However, [this] allows humans to focus on tasks that are most pertinent to our human skill set, rather than finding that needle in the haystack.
“If we can be more strategic with our decision making, we can get a better tempo on the battlefield against our enemies, and that will eventually save lives.”
