My first multirotor build was back in late spring of 2014: the Flying Cinema CineTank Mk1. I built it out of frustration with poor quality and jello-ridden GoPro footage I was getting using my Blade 350QX that I’d had since mid-2013. The CineTank promised relatively smooth footage thanks to its clean frame/dirty frame design and was getting a lot of kudos for innovation on the various multirotor forums. My CineTank kit arrived in May 2014 and was a lot of fun to build. I learned quite a bit about the engineering, integration, power distribution, camera placement, ESC quality, and flight controller setup during the two week assembly and tuning process. It was a great first build and I’d do it again without hesitation.
However, GoPro footage wasn’t as smooth as I liked, so when Flying Cinema announced their brushless 2-axis GoPro gimbal I was all over it. I’d been reading on FPVLabs and RCGroups where brushless gimbals were the bomb for silky video. Some very talented pilots and video editors posted excellent clips of buttery smooth flights over amazingly beautiful landscapes. As a photographer, I was sold.
So I transferred over all the flight control hardware over to my new CineTank MkII complete with 2-axis gimbal. This new setup got me to where I wanted to be in terms of the air vehicle component of my burgeoning aerial photography hobby and soon to be (dare I dream) business.
I found myself flying over water quite a bit at the time, which was when I lived in Virginia. At this point I’d only been in the hobby for a couple of years, so even though the MkII was my second build and I used top shelf components, quadcopters have zero survivability in the event of motor or ESC failure. Deep down I don’t think I trusted my own soldering skills.
As any of you readers who fly drones know, when the drone bug bites you it bites hard. So around November 2014, I decided to try something a little more unconventional and failure tolerant for my frequent overwater flights – the 3D Robotics Y6B hexacopter.
I only used the 3DR Y6B frame – everything else was upgraded. I put six oversized motors and high quality ESCs on this drone in addition to an M8N GPS that allowed me to lock upwards of 19 satellites during flight. I was flying a GoPro under the Y6B. Originally I used the Tarot 2D gimbal 3DR recommended, but quickly felt limited by the 2-axis design. At this point in my drone career, I was convinced that the GoPro Hero 4 was the key to amazing footage and photos, so I swapped gimbals to the Feiyu Tech Mini3D in order to have 3-axis camera flexibility under the Y6B.
At the time both of my aerial photography drones had onboard camera switches so I could change ground station views between the flight camera and the GoPro using a switch on my controller. This allowed me to frame the shots with the GoPro cameras remotely since using the built in wifi is a no-go with a 2.4 GHz control link and GPS (I learned that the hard way with the 350QX once back in 2013). I modified my Taranis controller with an extra thumbstick so I could better control the gimbal on the Y6B; everything worked pretty well from a control standpoint. I even went so far as to spend more than I care to mention to modify my GoPro with a flat rectilinear lens so there’d be no more fisheye. Just getting this new lens correctly focused took several hours, but I figured it was worth it to get the most out of the GoPro and the Y6B.
Despite all this hardware, I continually had problems with exposure, ISO, and white balance on images from the GoPro cameras, which as a photographer frustrated me to no end. No matter how hard I tried I couldn’t correct these problems during post to my satisfaction, especially with having no control over the camera shutter speed. I figured it was probably just my n00bishness at aerial photography, so I spent the next several months of summer and fall of 2015 honing my piloting and aerial photography skills.
I began exploring how to mount a better camera on my Y6B. I wanted something like a point and shoot that had the ability to geotag its photos and could be manually triggered using the functionality built into the Pixhawk flight controller. I also wanted to be able to see the shot framing from the ground station. I never found anything that was a plug and play option, so the idea was shelved.
Then in May 2015 3D Robotics announced the Solo quadcopter. The Solo promised integration of GoPro cameras into the UAS allowing full ground station control of all the things that had been frustrating me with my current setups. With the Y6B and the MkII, I had to start the GoPro on the ground, hope I got the settings right, and wait to see the end results during post. As you can see from the photo above, those results were frequently disappointing.
Also disappointing was the fact there didn’t appear to be any way to get the same level of camera integration into the drone that was offered by the Solo. I even asked 3D Robotics on Twitter if the Solo gimbal would ever be compatable with the rest of their offerings. 3DR’s response: “It’s a Solo-specific controller. Might work with products to come, but not older ones”.
One of the podcasts I subscribed to back then was Multirotor Podcast. I learned a lot from listening to Joe Papa and Erick Royer talk about all things aerial photography. Joe was particularly passionate about the craft and could sell proscuitto to a rabbi, as the saying goes. My wife even got into the vibe once while we were listening to the podcast on our way out to Lexington for a family getaway. Joe was passionately extolling the virtues (as he does) of the ImmersionRC Vortex FPV racer. My wife excitedly said, “You should get one of those!” Twist my arm. Even had it sent next day air.
Anyway, Joe and Erick had both been consistently giving the DJI Inspire 1 solid marks since February 2015. Joe went pretty far with his praise in early September 2015 when DJI announced the X5 camera for the Inspire. He said that with the new camera, this copter was able to do 99% of the work asked of aerial photographers and was now his go-to air vehicle for jobs. The overall system integration on the Inspire was always excellent and now the X5’s micro 4/3 sensor was consistent with professional aerial photography industry standards. Erick asked about reliability concerns since it was just a quadcopter, and Joe (passionate as always) said that after dozens of flight hours on the Inspire, he has zero concerns about reliability with the drone.
Unfortunately Joe and Erick parted ways in early 2016. Erick started doing the Drone Vibes Podcast with aerial photographer Petr Hejl and Joe was doing a new podcast called DroneLife. Erick and Petr ceased delivering content on December 1st of 2016, which was disappointing as I learned a lot from Petr. Joe’s solo podcast was awesome but sputtered out after only 10 episodes.
At any rate, in late summer 2015 the Inspire was rapidly climbing to the top of my wish list, despite my misgivings with DJI.
My Inspire 1 showed up on September 28th, 2015.
Grudgingly I had to hand it to DJI. The Inspire is engineered extremely well and the quality of integration between the flight controller, camera, remote controller, and the rest of the system is simply incredible. I’ve always thought the Inspire was a neat-looking design and was pleased with the user experience when a friend at my old flying club back in Virginia let me fly his late last summer. Joe Papa’s compelling reviews and my growing GoPro frustration pushed me over the edge.
Over the fall of 2015 as I gained a deeper level of understanding of this UAS, I was continually impressed with it as I learned more and tested different applications. The Inspire 1 geotags all the photos and offers full camera control from the ground station – ISO, shutter speed, white balance, EV comp, everything. I began experimenting with 3D modeling and orthorectification using Agisoft Photoscan and some of the SDK-based GCS apps for the Inspire like Maps Made Easy, DroneDeploy, and Autopilot. The best part of all this is that the Inspire flies so smoothly and predictably that I could finally concentrate on being a photographer. It was a very relaxing flight experience for me.
The interchangable payload design of the Inspire was also very compelling, particularly since as far back as fall 2015 I knew an aerial photography business was in my future.
Meanwhile, the integration with DIY UAS like my Y6B and MkII seemed to have hit a wall. About the best thing out there integration-wise was the Amimon Connex which sends nearly latency-free 1080P video up to 1500m away as well as overlay MAVLINK telemetry. Even with this setup, I’d still have no camera control and I’d be stuck with a GoPro.
This state of things is frustrating since I love building and tinkering, but there doesn’t appear to be any way to meet my integration requirements at the moment. My hope was that 3DR would eventually release a standalone GoPro gimbal and Pixhawk 2 that led to better integration, but they gave up on the prosumer market. The only Solo with a professional camera is the Solo Enterprise, but it is only sold with a $12,000 package that I don’t need, not to mention that I’m not even sure 3DR will be in business a year from now. I love my MkII and Y6B, but they’re trapped in 2013 technology – virtual antiques in the world of drones.
So as I sit here writing this, the workhorses of Talon Six Aerial will continue to be DJI multirotor drones. The MkII and Y6B sit on my shelf down in the workshop….relics of a bygone age.