Technology executive says the next generation of autonomous systems will be defined by how well they operate when GPS, communications, and cloud connectivity cannot be guaranteed.

WASHINGTON, D.C. / ACCESS Newswire / September 16, 2026 / As robots move out of controlled settings and into mines, industrial facilities, infrastructure sites, and defense environments, technology executive Brandon Torres Declet believes the industry needs to reconsider one of its most common assumptions: that connectivity will always be available.

Cloud computing has transformed modern technology, but autonomous machines increasingly operate in places where sending information to a remote system and waiting for instructions may not be practical. Underground environments can block GPS and communications. Buildings and industrial sites can create connectivity problems. Defense operations may take place where signals are disrupted or intentionally denied.

According to Torres Declet, these environments are pushing robotics toward greater use of edge autonomy, where critical computing and decision-making happen onboard the machine.

“The cloud is incredibly useful, but a robot operating in the physical world cannot assume that it will always be able to reach it,” said Torres Declet. “If communications disappear or GPS becomes unavailable, the machine still has to understand enough about its surroundings to make a sensible decision. That could mean continuing the mission, changing its behavior, returning to a known location, or recognizing that it should stop.”

Torres Declet has spent much of his career working across drones, autonomous systems, aviation, and emerging technology. His experience includes commercializing drone services and software, as well as work on autonomous aerial and ground systems designed for difficult operating environments.

That experience has made him particularly interested in what happens when autonomous technology leaves a controlled demonstration and encounters the uncertainty of the real world.

Moving Critical Decisions Closer to the Robot

Many connected systems can send information elsewhere for processing and receive a response almost instantly. For autonomous robots, however, even a short delay can matter when a machine moves through an unfamiliar environment.

Edge autonomy moves certain decisions onto the robot itself. Depending on the system and mission, onboard computing can let a machine process sensor information, understand its location, recognize obstacles, update maps, and decide how to move without constant communication with an outside network.

Torres Declet believes the important question is not whether every function should move away from the cloud. Instead, robotics developers need to determine which decisions cannot afford to depend on connectivity.

“I do not think this is an argument between the edge and the cloud,” Torres Declet said. “There are things the cloud does extremely well, and that is not going to change. The engineering question is what the robot absolutely needs to be capable of doing on its own when the connection is unavailable.”

That distinction becomes increasingly important as robots are sent into environments that are difficult, dangerous, or impractical for people to enter.

Designing Autonomous Systems for Imperfect Conditions

Torres Declet also cautions against treating autonomy as a question of whether a robot can simply continue operating when something goes wrong. In some situations, continuing may be exactly the wrong decision.

A machine may lose confidence in its location. A sensor may begin producing unreliable information. Conditions may change in a way the system did not anticipate. Good autonomous behavior may require the robot to slow down, alter its route, return, or stop rather than blindly continue its original mission.

“Resilience does not mean pretending failure cannot happen,” Torres Declet said. “It means thinking seriously about how the system behaves when it does. I want to know what happens when a sensor is unreliable, when communications drop, or when the environment is different from what the machine expected. Those situations tell you a great deal about how useful the technology will be outside a controlled setting.”

Building that capability involves tradeoffs. More onboard computing can require additional power and may affect weight, heat, cost, and system complexity. Those considerations can matter most for aerial robots, where every component affects payload capacity and flight time.

For Torres Declet, those constraints are part of the engineering challenge, not reasons to avoid edge autonomy. Different robots and missions will require different balances between onboard and remote computing.

Robotics Is Moving Into Harder Environments

The shift matters because the environments where autonomous machines can provide significant value are often the same environments where reliable connectivity is hardest to guarantee.

An autonomous system inspecting an underground mine faces different conditions from a robot operating in a controlled warehouse. A machine mapping an unfamiliar building or supporting a defense mission may face an entirely different set of constraints.

As those applications expand, Torres Declet expects resilience to become a more important measure of autonomous technology.

“We spend a lot of time asking what robots can do, and we should,” he said. “But as these systems take on harder jobs, we also have to ask what they can still do when part of the environment stops cooperating. That is where autonomy starts to become much more meaningful.”

The future, in his view, is unlikely to be entirely cloud-based or entirely independent of networks. Instead, autonomous systems will increasingly divide responsibilities between onboard intelligence and broader connected infrastructure.

For robotics companies, deciding where that boundary belongs may become one of the industry’s defining technical challenges.

About Brandon Torres Declet

Brandon Torres Declet is a technology executive, founder, and board director with more than 20 years of experience across autonomy, robotics, drones, aviation, defense technology, and emerging technology commercialization. His career includes leadership roles at private and public companies, as well as experience in M&A, capital markets, corporate strategy, and regulatory affairs.

Torres Declet founded MEASURE, an aerial intelligence company that developed drone services and the Ground Control software platform. He has also held executive and board leadership roles across the drone and robotics industries. His current work includes physical AI, edge autonomy, GPS-denied navigation, autonomous mapping, and applying autonomous systems in commercial and national-security environments.

Earlier in his career, Torres Declet worked in national security with the NYPD Counter Terrorism Bureau and the U.S. Congress. He holds a B.A. from Union College, a J.D. from Fordham University School of Law, and an LL.M. in National Security from Georgetown University Law Center.

Media Contact

https://www.brandondeclet.com/
info@brandondeclet.com

SOURCE: Brandon Declet

View the original press release on ACCESS Newswire

Media gallery

About The Author
Brandon Declet