Tenbo Air

     Last week, the Federal Aviation Administration (FAA) issued a press release announcing a proposed list of regulations for users of drones and Unmanned Aerial Vehicles (UAVs). According to the statement, these regulations are intended to “allow routine use of certain small unmanned aircraft systems (UAS) in today’s aviation system, while maintaining flexibility to accommodate future technological innovations.” Small UAV usage has increased substantially over the past few years, but the FAA has done little to regulate the safe operation of these systems despite mounting pressure from advocates ranging from commercial users to the National Parks Service to corporations like Amazon and Google. Even President Obama called on the FAA to expedite the release of their regulations after a late-night incident in which a secret service agent, following one too many drinks, crashed his DJI Phantom on the White House lawn. In a 2012 hearing, Congress gave the FAA a deadline of September 30, 2015 to come up with a plan for safe integration of remote controlled aircrafts.

Aftermath of the DJI Phantom which crashed on the lawn of the White House January 26th, 2015

     Although government regulation may hinder creativity and certain possibilities of operation, a legal framework is necessary for the safe growth of the UAV industry. As drone technology becomes cheaper and more widely available, the market for it expands beyond professional use. Amateur fliers often don’t take necessary precautions like checking for tight propellers and obeying a flight ceiling of 400 feet. Despite preexisting rules stating that UAVS must remain at least 3 miles away from airports, there have been instances both of UAV sightings near commercial aircraft and of narrowly avoided midair collisions with helicopters flying higher than 1,000 feet. Though many camera-equipped drones come ready to fly straight out of the box, it takes hours of practice and training before operators can reasonably consider themselves skilled. Government regulations will promote a higher standard of safety and professionalism by ensuring that this technology is not used in reckless or dangerous ways. The FAA’s proposed regulations, however, have been widely critiqued since their release.

You can find a full overview of proposed rules on the FAA website, but we’ll summarize a few of them here:

Regulations would require that all remotely piloted and unmanned aircraft must weigh less than 55 pounds, not exceed 100 mph airspeed or an altitude of 500 feet, and not be flown at night. 

Individual systems would not need to be FAA airworthiness certified, although operators would be required to make sure that their crafts are properly marked and in safe condition before every flight. 

Another new rule states that an operator would not need a private pilot’s license to fly a drone, even though the few businesses that have already been approved for drone flight by the FAA were required to send their operators to flight school. Under the proposed rules, operators must “pass an initial aeronautical knowledge test” at an FAA-approved center every two years and must obtain an unmanned aircraft operator certificate. Obtaining a private pilot’s license would be much more expensive than this “aeronautical knowledge test,” and would focus on manned airplane flight protocol, so these new requirements would be more appropriate and helpful for educating users about UAV-specific issues.

Of all the FAA’s proposed regulations, the most contentious and potentially damaging is its Visual Line of Sight (VLOS) requirement, which would require that an aircraft remain within sight of either the operator or a visual observer at all times. Although this requirement might seem like an obvious one, it essentially eliminates one of the main reasons for utilizing drones over other means of aerial data capture. Although the intention of the rule is to prevent amateur operators from relying solely on First Person View (FPV) to monitor the position of their craft, an unsafe practice even for an expert pilot, it has unintended negative impacts for the future of the industry. Requiring that an aircraft remain visible to the operator contradicts most uses for automation. For example, Amazon has been planning on using drones to deliver packages for quite some time, but this new regulation would require each aircraft to be followed by an operator, making the entire service no more efficient than current delivery methods. Agricultural drones, intended to monitor crop health for huge areas of land, could not legally operate if the area were forested and the line of sight blocked. Hobbyists who use their UAVs to explore inaccessible areas would have to literally lower their sights. Almost every drone is equipped with basic failsafe mechanisms that prevent the aircraft from falling out of the sky, and via the on-board GPS an operator can see where his drone is located without having to visually monitor it. Although the VLOS rule is intended to keep operators and spectators safe, it drastically reduces the functionality and usefulness of autonomous flying, and has already caused large companies such as Amazon Prime Air to take their jobs overseas.

How we made this model

Beloved by locals, this small castle in the Irish Bayou provides a memorable landmark when crossing Lake Pontchartrain between New Orleans and Slidell. Officially known as Fisherman’s castle, it was built in 1981 as an attraction for visitors to New Orleans during the 1984 World’s Fair. The 942 square-foot structure was built to resemble a 14th century chateau, and although it is located right on the banks of the Irish Bayou, the castle has survived many hurricanes.

To create this 3D model, we flew a quadcopter in concentric rings around the castle and took photos at predetermined intervals. This concentric ring pattern is crucial for creating an overlap between images, which gives texture and depth to the model. Our photos focused on the castle, so the house in the background appears warped. Traditionally, such a model could only be made using LiDAR technology, which is more expensive and prohibitively heavy for use on a drone, but new advances in computer processing technology have closed the gap between LiDAR and traditional photography for generating 3D models. These models can be produced in an astonishingly short period of time, requiring only several hours of work from a single person.  

For what is 3D modeling useful?

3D modeling is already popular in the fields of archaeology and historical preservation. Cultural sites are always at risk of damage and destruction due to anthropogenic factors like urban development, pollution, and war, as well as natural disasters such as hurricanes and earthquakes. Using drones and this 3D modeling technology, the intricate details of large historical structures such as plantations and forts can be preserved and represented in interactive museum displays. Archaeological sites can be quickly excavated, catalogued, and photographed, so that experts can analyze the distribution of artifacts at a later date. This process expedites the collection of data, and provides a medium for individuals with differing skills to evaluate sites and provide new insight for drawing conclusions. According to section 106 of the National Historic Preservation Act, any project funded by the federal government  must take into account any impact that the project will have on sites listed, or eligible to be listed, on the National Register of Historic Places. Often this includes an archaeological excavation, which can take weeks or months to complete. By using 3D modeling, the time needed to gather and analyze cultural sites can be drastically minimized, and sites that will inevitably be impacted by new construction can have these damages mitigated through digital preservation of the structure. In a culturally rich city such as New Orleans, this technology is critical for conserving and curating our history for future generations. 3D modeling expedites new development while simultaneously encouraging the preservation and display of cultural sites. This technology can and should be used to create a record of the beautiful structures that set New Orleans apart from the rest of the world.

Tenbo was created in 2014 by the three founders of successful solar power installation startup, Joule.  Since our inception, we have been committed to researching and developing the equipment necessary to enter the commercial drone industry.  At Tenbo, which translates to perspective in Japanese, our primary objective is to use drones in order to collect aerial imagery that will provide end users with an overhead view of their project sites.  We then analyze and process collected data into finalized maps, videos, 3D models, and more.  Based out of New Orleans, Louisiana, we are one of the only companies in the southeastern United States using drones for surveying, mapping, modeling, photography, and videography.

Although a newer organization, Tenbo has already been a part of some very interesting projects, such as mapping a portion of recent construction for the Make It Right Project in New Orleans’ Lower Ninth Ward. Not only can we offer immediately up to date data that cannot be seen on services like Google Earth, but we can do so at a much higher resolution than traditional plane mapping or satellite imagery. For example, we mapped a section of the Lower Ninth Ward at a stunningly clear 1” per pixel.

When Make It Right was shown our images, they immediately pointed out a solar panel whose replacement had been neglected. This is a simple example of just how immediately useful a high resolution 2D image can be.

Where are we headed?

There are many industrial uses for commercial drones, but we are currently most excited about aerial mapping, especially for surveying purposes. Traditional surveying requires multiple people and several days to generate hundreds of GPS points in a grid formation across an area. Using only tens of ground control points and one of our drones, we can gather millions of data points across hundreds of acres with similar accuracy (1-2 inches) in as little as 2 hours, with a single operator. Photogrammetric data eliminates the need to guesstimate the position of the terrain between our control points, therefore our data is more complete than traditional surveying.

We use camera images instead of LiDAR or a traditional GPS point grid, allowing for faster, cheaper processing and resulting in data comparable in accuracy and quality. Advances in software and processing power have revolutionized the industry; these innovations will soon become the standard for surveying and mapping.