As drone technology becomes increasingly automated, the level of human involvement is shifting from remote pilots in the field to remote operators in the office. This approach can enable more scalability and operational oversight as enterprises grow their drone fleets to inspect pipelines, monitor crops, or survey infrastructure. However, several barriers still exist when it comes to safely enabling Beyond Visual Line of Sight (BVLOS) operations.
Let’s explore some of the things that are required to unlock BVLOS operations, including remote automation, safety & compliance rule enforcement, AI-powered cybersecurity, and more.
Drones are disrupting various industries and innovating outdated business models. BVLOS enables UAVs to operate beyond the normal vision range of the pilot. BVLOS capabilities are becoming a quintessential aspect of the drone industry. They provide numerous benefits over the regular line of sight flights. They are cost-effective, energy-efficient with fewer takeoffs and landing phases, cover significant ground in a single flight, and drones’ low-altitude flying capability can help in high-resolution data collection.
In many cases, businesses need to operate drones beyond visual line of sight to complete a wide range of missions, such as assessing hurricane damage and delivering aid to the devastated areas or inspecting pipelines to prevent leaks in the oil and gas industry. That means they’ll need more advanced technology in place to identify other aircraft, stay up to date on airspace changes, and safely reroute drones to avoid potential hazards. Let’s first consider the challenges commercial drone operators are facing today.
We all know there are many benefits to launching a drone operation, but navigating low-altitude airspace is complex. The burden has fallen on drone operators to manually evaluate the airspace, plan their flight paths, and avoid hazards as conditions change. But this approach isn’t scalable when you consider the volume of data operators are expected to evaluate for a successful mission.
Today’s systems require too many manual workflows that limit scalability and leave room for error in the rapidly changing airspace. Drone operators are expected to monitor weather changes, avoid buildings and construction cranes, factor in risks on the ground, and comply with shifting regulatory dynamics. The burden typically falls on them to manually plan, execute, and adapt their flights as these conditions change.
At the same, drone operators are challenged by disconnected systems. They typically have to use several different tools to check airspace conditions, plan and execute missions, and gather insights. But it’s a cumbersome process that leads to disconnected information as operators switch between different applications.
- Traffic: For starters, operators need to check airspace traffic, including both manned and unmanned traffic, to maintain safe separation. To minimize public safety risks, operators also need to evaluate activity on the ground below, such as roadway and foot traffic.
- Regulations: From a regulatory standpoint, they also need to check airspace classes and boundaries and monitor shifting dynamics, such as temporary flight restrictions and notices to airmen.
- Weather: Access to micro-weather data is also important to check precipitation, wind, temperature, and visibility. These factors can impact the flight path, battery life, and overall success of the mission.
- Infrastructure: Drone operators also need to evaluate local buildings, bridges, schools, stadiums, and airports to navigate around densely populated areas.
- Environment: They also must also check the local elevation and terrain to avoid potential hazards.
Security and safety
Those are just the external factors operators are expected to evaluate. Operators also have to consider the health and security of their aircraft. They’re ultimately responsible for protecting their drones from both intentional acts, such as cyber threats, and unintentional acts, such as hardware malfunction.
- Aircraft health: This requires operators to continuously monitor their vehicle health, but that becomes a lot more challenging as a fleet grows.
- Aircraft security: From a security perspective, operators also are expected to protect their drones from malicious activity. Just like the computers we use today, drones can be hacked if not appropriately secured, posing dangers to people and property on the ground.
Automation, AI & Blockchain
The bottom line is it’s not feasible to manually monitor and interpret this exceptional volume of data at scale. A new approach is required to simplify drone operations with a connected system that automates every phase of flight, removes the burden on drone operators, and allows operators to focus on overseeing the mission’s success. This is where advanced technologies like artificial intelligence and blockchain can help.
AI algorithms are trained to analyze a large volume, variety, and velocity of data and instantly act on the insights. These algorithms automatically learn from patterns to uncover and act on trends hidden from the human eye.
In technical terms, blockchain is a distributed ledger of immutable records stored in a decentralized database. In layman’s terms, it enables safe and accurate record-keeping across a network of computers, allowing multiple parties to interact with the same universal source of truth using a private key. “Smart contracts” are also a key component of blockchain technology. Smart contracts can be encoded on any blockchain to set rules mutually agreed upon by network members and automatically execute the terms without human intervention.
When used in parallel, these advanced technologies can help eliminate manual workflows and enable safe BVLOS operations. Let’s walk you through a few examples.
AI algorithms can be trained to calculate the optimal route for one or more drones based on the mission parameters, such as the start and endpoint, desired cruise altitude, timeframe, and payload details. These algorithms can also factor in airspace, vehicle, and location data, such as weather, terrain, population density, and roadway traffic, to generate routes that minimize risks in the air and on the ground.
During the flight, the AI models will monitor, predict, and adapt to conditions as they change. This approach essentially removes the burden on commercial operators by enabling autonomous workflows that are safe and scalable as a fleet grows.
When it comes to regulatory compliance, blockchain augmented with smart contracts can encode the airspace rules, such as flying below 400 feet during daylight hours, as mandatory parameters in a flight planning system. Organizations can also use this technology to set additional company-wide safety standards for their commercial drone operations, such as flying with at least 20% battery life under 25 mph winds.
The blockchain smart contracts automatically record information onto the ledger and execute the terms without human intervention. This approach helps automate compliance with the rules before flight authorization and during flight as airspace conditions change. It also helps ensure all drone operators associated with your organization are following the same rulebook.
When it comes to monitoring your vehicle health, predictive AI technology can remove the burden on operators by analyzing sensor data across your fleet and flagging suboptimal operations. An AI-based approach can more accurately monitor performance to forecast vehicle health and identify impending failures before they occur.
If a potential issue is identified, such as a degrading battery, AI technology can automatically generate a maintenance request and assign the request to a technician upon landing at a facility. Blockchain technology, augmented with smart contracts, can also ensure the maintenance request is resolved and signed off by a technician’s private key before the drone can operate again.
In the emerging UAV environment, new security threats will often take the form of previously unseen, “zero-day” attacks. Traditional anti-malware software, dependent on signatures of known threats, won’t be adequate to detect this unknown malware. AI-powered cybersecurity will be critical to detect malicious activity on the edge and prevent it from executing on a drone.
An AI-based approach can learn the DNA of what a malicious file might look like instead of relying on an existing threat database. This approach protects drones from never-before-seen attacks and can still function when network connectivity is non-existent or impaired.
Systems today are largely disconnected and still rely on humans to manually plan their flights, comply with regulations, and adapt to changing conditions. Advanced technologies like AI and blockchain can enable a new, automated approach. This approach still relies on human input, but it allows more scalability by automatically planning, executing, and adapting flights as conditions change. It also enables enterprises to scale their drone operations by ensuring all pilots associated with their organization remain compliant with the regulations, business rules, and safety standards.