I started down this path with an interest in 3D modeling of caves. I was experimenting using a kinect to build a 3d model but it was clunky and unwieldy. I ran across the term “structure from motion”… which just means that you can recreate 3D structure from the motion of a camera through a scene – the practical result of which is that you can create a 3D model from photographs. I am convinced thoroughly that this method should become widespread in all sorts of digital collections. I presented this year at the Society for American Archaeology (SAA) Annual Meeting on creating models which are measurable comparable to the real world measurements. This process is simple to implement and the photographic techniques required relatively simple – but the benefit is rich 3D content which can be shared or recreated in a multitude of ways.

This is not limited, of course, to archaeology. This can be used to share art, architecture or landscape content – anything really. The NPR show Radiolab had a segment about a paleoarchaeological find of a human ancestor which was accompanied by a 3D model which was intended in the absolute best case scenario to be 3D printed and held while listening to the episode. This is an incredible feat – one of the first in which an extraordinarily rare find was made so accessible through the use of 3D scanning. 3D structure from motion means that no longer do you need a laser scanner – the model I created for the SAA meeting and that I will attach here was created with nothing but a camera and some computer time. Agisoft Photoscan is incredibly good at this point. As part of my work as a student innovation fellow I have been developing lesson plans – one centered around accurate terrain modeling from UAS photos and the other based on modeling objects with handheld photographs – which I will post here once I finish editing some of my stream of consciousness out of them – and make some better figures.

Some of the most recent beneficial developments in Photoscan have been improvements in the auto-detecting markers that Photoscan can produce. Experiments with this system so far have shown it to be pretty reliable in detecting the markers correctly which greatly simplifies the creation of accurate, measurable models.

I find it fitting that as I wind down the semester and look towards my next steps I find I have been invited to, finally, tackle the problem which initially led me down this path of discovery with regards to 3D modeling – caves. During and then after my undergraduate degree I worked at a site with a number of caves known as Pacbitun. I will be returning there this summer to assist with creating 3D models of the very same caves that got me interested in this realm.

This is all to say nothing of the other major avenue I have been pursuing as a Student Innovation Fellow – Unmanned Aerial Systems.

So, it has been well over one year now since I was first exposed to the use of unmanned systems in archaeology. Where are we right now, and where are we going?

This year saw the first dedicated session at the Society for American Archaeology Annual Meeting, chaired by Dr. Gerardo Gutierrez of the University of Colorado, Boulder.

The papers in that session made one thing clear, one after another – Unmanned Aerial Systems – yes drones – are going to be a part of the future in archaeology. They will revolutionize the field in many ways, allowing access to unprecedented levels of detail and control in aerial data collection. However the biggest lesson I took away from my work is that – despite the success had drones are not the answer much of the time. It is often easier, safer and more cost and time effective to collect imagery using handheld cameras and perhaps some creative photography gear (like gibs or “selfie pole” style sticks) rather than using a drone.

When you do cross over into needing an unmanned aerial system there are a number of routes to take – DJI offers an easy to use product that nevertheless suffers from some persistent flaws. Better autopilots exist but at the expense of needing an expert to set up and maintain the aircraft. There are also very very expensive “automatic” systems which are heavily computer controlled and as a result can be operated by nearly anyone… and did I mention VERY expensive?

So, when should you use a drone? Fixed wing aircraft excel in many respects but are limited by takeoff and landing requirements. Multi-rotors suffer from a number of limitations, chiefly battery life, but excel in maneuverability. Anywhere that takeoff and landing space is limited is a good candidate for using a multi-rotor system. Both of these systems are limited by the skill of the pilot in the terrain provided. In a dense enclosed environment where most of the area of interest is beneath the canopy, using unmanned systems in general may be unnecessary or even dangerous.

Moving forward, unmanned systems will continue to see increasing use in archaeology and related fields. This will coincide with increasing use of these systems in other fields, including geosciences and agriculture. As the FAA begins certifying pilots and systems for commercial and research use in the US drone based project will be come more and more common.