As power utilities continue to modernize operations, many companies are in the process of building drone programs to replace manual asset monitoring and inspections. It’s a no-regrets move, and drone capabilities have been proven to enhance reporting accuracy, increase efficiency, and improve worker safety. But just as utilities are starting to catch up to the latest tech, another important evolution is underway – fully autonomous drones, with the computing capacity to see, understand, and react in real time without human involvement. In other words, those “eyes in the skies” are about to get AI-powered brains.

THE NEXT GENERATION OF DRONES

Currently, utilities are deploying drones equipped with high-resolution cameras, LiDAR sensors, and thermal imaging technology that can provide critical knowledge on the health of their assets. From there, drone images are analyzed by AI to generate insights that enable proactive/predictive maintenance, improve forecasting models, and ultimately reduce site visits and power outages.

This capability has been transformative, but it can also be limiting – because in addition to the upfront investment of the equipment and AI software, most drones also require a trained pilot. That’s led many forward-looking utilities to consider ways of autonomizing drone programs (i.e. no human pilot required) to make drones more efficient, intuitive, and perceptive. It’s early days, but recent progress has moved the industry past proof-of-concept stages and into experimentation, signaling a breakout moment ahead.

One of the main differences with autonomous drones will be the AI-driven navigation with built-in obstacle avoidance technology, allowing drones to recognize (and avoid) trees, poles, and branches in real time and operate in complex environments with complete awareness of their surroundings. Autonomous drones will also be equipped with AI-powered defect detection, which can accurately identify structural issues, corrosion, vegetation encroachment, and electrical defects/anomalies without exposing workers to hazardous environments.

HOW AUTONOMY CAN HELP ADDRESS COMMON DRONE PROGRAM CHALLENGES

Before we get into any futuristic potential, it’s important to recognize the tremendous progress that utilities have already made updating their equipment and processes over the past few years. As industrial equipment ages, weather events intensify, and electricity demand surges, utilities are under constant pressure to maintain the power grid and reduce customer downtimes, often with limited budgets.

Each utility is unique in its size, territory, and available resources, but for those utilities that are further along in their drone programs, it can be helpful to compare existing capabilities to future potential. The following section outlines common challenges utilities face in scaling drone programs as part of their larger inspection modernization efforts, and how autonomous systems are designed to help.

1. Technical Limitations

Power utilities are responsible for inspecting and maintaining hundreds of miles of T&D lines snaking through forests, cities, and rural areas. Servicing extensive and/or remote territories can be difficult, even with drones. Battery life constraints mean limited flight durations that can restrict coverage area per deployment, and all that new technology loaded onto the drone (like LiDAR and other sensors) adds weight that can reduce flight efficiency.

One of the more exciting developments with autonomous drones is that they come with autonomous charging docks, where drones can land, recharge, and relaunch without human intervention. These drones are also being outfitted with longer-lasting, lightweight batteries that can extend flight time, and AI-driven path planning that enhances navigation capabilities through predictive route optimization. By bringing these capabilities on board, autonomous drones can execute continuous inspections with little to no oversight.

2. Data Management and Analysis

Acquiring data is just one aspect of an effective drone program, and as drones gather more information from the field, that data also needs to be properly processed and stored. Any hope of AI-powered insights depends on algorithms that can accurately identify and categorize defects to minimize false positives/negatives, and once those insights are formed, they need to be fed into existing asset management platforms for effective decision-making.

The good news is that autonomous drones can help expedite data management with AI models deployed on the drone itself that use edge computing for real-time processing and faster defect identification. Autonomous drones will also seamlessly integrate with utility servers, allowing for secure data transmissions that ensure efficient and protected data processing. And with automated AI-driven insights, machine learning algorithms can process drone-collected imagery and automatically detect anomalies, reducing the burden on human analysts.

3. Security

Drones currently rely on cloud-based AI and remote communications, making them juicy targets for hacking. Protecting drone-collected data from tampering or unauthorized access is a key part of a utility’s risk management strategy, and data must be handled securely in accordance with industry and government standards to remain in compliance with regulators. Oversight of this process can be time consuming, and mistakes can be costly.

Autonomous drones aim to alleviate this burden by implementing comprehensive AI-based cybersecurity solutions, like Mobilicom's ICE Cybersecurity Suite, which autonomously detects, prevents, and responds to multiple cyber threats in real-time without requiring operator intervention. All autonomous drone-captured data would also be encrypted and transmitted securely over networks like Starlink or utility-approved private networks, helping prevent unauthorized access and ensuring drones cannot be overtaken by rogue entities. As the technology continues to mature, developers are collaborating with agencies to create operational standards for autonomous drone flights and form regulatory protocols to guide widespread industry adoption.

THE IMPENDING DECLARATION OF DRONE INDEPENDENCE

There is no doubt that autonomous drones will one day be the standard for utility asset inspection, and companies like Skydio, Percepto, and Exyn Technologies have emerged as early leaders in the field. Signs of progress are everywhere, including a recent project between Skydio and Firmatek using autonomous drones that reduced Firmatek’s pole inspection time from 30 minutes per pole to 30 seconds, equating to a 92% total time savings that decreased their project timeline from three months to just one week.

As more industry case studies emerge over the coming months, remember that autonomous capabilities are not just features or add-ons to an existing drone program. Rather, autonomy needs to be baked into drone architecture from the beginning, which presents a unique opportunity for utilities just getting started on their modernization journey. By focusing on updating core processes now and staying apprised of emerging technology for the future, utilities can help ensure their assets (and their company) will stand the test of time.