Drones

For the past 10 years or so, I have been playing around with drones. I have to say this habit started much earlier when as a 5 year old, my dad and I build balsa wood & tissue rubber band airplanes. We progressed towards RC airplanes. We never got too far during my childhood, but the itch to fly was berthed deep within me. My dad and I would travel to airshows to see aircraft flying and static displays I could climb in. I took flying lessons in a Cessna when I was 16, dreaming about becoming an airline pilot, or even better, a fighter jet pilot. But life’s twists and turns caused my aspirations to shift as I moved towards college and career. Once I had my first job, I got back into RC aircraft, actually helicopters. I bought a Blade CX which is a coaxial/cyclical helicopter that was fairly easy to fly around the house. I always thought quad rotors were silly as they didn’t parallel actual human rated aircraft, but my mind was changed when I started flying around a Dromida Ominus about 4 years ago.

With Quidzel, I also have been working on some drone developments looking for business use cases. Recently I got my hands on a DJI Phantom 3, which blew me away with the software capabilities and refinement. I soon was watching videos of a new to market folding drone, the DJI Mavic, and I had to get my hands on it. This thing is incredible, and along with that, I recently picked up the FAA part 107 certification for commercial flying. With the current restrictions on autonomous drone flight and maintaining line of sight to a human operator, I think the use cases may be restricted to photography and other imaging applications. While everyone is picking up real estate photography and drone videos for weddings, I see a better value use case in inspections or even possibly remote manipulation.

I’m currently exploring ideas in the drone space, as I think this market is set to explode. While a lot of commercial drone operators are opting for the DJI Inspire platform, I think the Mavic is actually a better fit for many cases. It is less invasive, both visually and acoustically. Because of the light weight, it will do less damage to itself or other things in the event of a mishap. It has all the latest DJI technologies, including a very robust ISM datalink, forward & downward sensors, and an impressive battery life for this class of aerial platform. Its compact nature allows easy fielding and setup.

I have really felt bad for the GoPro Karma. They released their drone a week before the Mavic, and they were first to market with a practical folding design, built in display, and an intuitive removable gimbal for their product line. But the Mavic came out at 1/4 the folded size, and revealed GoPro’s failures to stay close to DJI in terms of technology. Initial reviews have panned the Karma for lack of downward sonar/cameras that can keep the drone perfectly still, the short flight times, noisy operation, and basically antiquated drone technology. I hope that GoPro can stick with it and release a 2.0 that builds on the failures of Karma 1.0, otherwise DJI will continue to dominate the prosumer drone market. I feel that GoPro is best positioned to effectively provide meaningful US-based competition for DJI, but for now, DJI has nearly the whole market around its pinky finger.

So, where does this leave me? I’m exploring new business cases that either involve software, hardware, or even just local aerial imaging/inspection services in the Chicagoland & Midwest. On the software front, DJI has an SDK that allows 3rd parties to build on top of their platforms. I’ve downloaded the SDK and I’m considering building some software that leverages the sensors to provide a dimensioned grid overlay to the video stream to enable measurement of findings, for drone based inspections. This allows repeatable measurements over time to track the status of a defect or other dynamic finding (plant growth, for another example). Automated image capture and automated flight control has some interesting possibilities as well. For example, I can develop 3 dimensional structural models with detailed imagery – i.e. fly 2m from a planar surface at 1 m/s, take image every 2 meters to create model that has a very fine visual granularity. I would process the images to eliminate known distortions, stitch together, etc to complete the model.

And what about the aux inputs? I’ve been exploring a transition here for a while. When I had a 1999 e46, it was my own problem I wanted to solve. But now I have kids and a newer car that has all these features now. I was always hoping to get a bluetooth device to market, but I kept running into challenges there. Right now, I haven’t spent a lot of time developing the business, but produce inventory, process orders, and handle customer service needs. I love helping people out, and see a future for this specific business. I think Quidzel’s aux input business needs to focus on the CD changer emulators for an expanded # of makes. There’s a lot of CD changer emulators out there for various makes, but I still like the intuitive/simple/low cost approach that Quidzel’s hardware offers. The fact is that I know and can support this for my customers vs. some place that doesn’t know what’s inside the box. I’d like to build a USB interface on it, allowing for remote end user firmware updates (current method is mailing reprogrammed modules around). There’s other functions that this device could do in terms of providing a simple USB interface to a number of 1 wire serial busses used on cars.

Anyways, I’m interested in your feedback, and thanks for reading my thoughts!

Posted in Uncategorized