I have been getting qoutes for a drone to add to our inspection business. So far I have quotes from Aibotix, Ascending Technologies, Aeryon and Altus. We currently use a Draganflyer X4. These professional drones have a massive price tag attached to them and I am wondering how they can be so expensive compared to a cinematoghraphy drone thats carrying a red epic camera around a movie set. What makes them so expensive? Whats in there that can add up to 65K. If you ask the manufacturers you get the same answer. There industrial grade, there not mass produced in china, there safer more reliable. yada yada what separates a 65k dollar drone from a drone that can be built with the best motors and best ESC's on a solid platform. Is it the flight controller? Is the flight controller in a falcon x8 or an Altus what separates these drones from the rest? I was wondering if someone could shed some light on this for me.
You might want to check with Aeronavics in New Zealand to see what they could put together. They have some interesting very high quality equipment. They seem to be moving more an more in to industrial areas and have a very good reputation.
Im a big pixhawk fan, I build systems for commercial & industrial clients, and also build/service/fly Altas, and know Aeronavics pretty well! Have yet to have the pleasure of playing with one of the Shotovers, I too want to see what $120k gets you!
Ill throw my 2c in -
Altas are fantastic machines, the quality is incredible, BUT the closed source of the FC isnt any good for inspection or any serious work other than filming. Im not saying the synapse isnt a nice FC, it is, but it just isnt designed for doing anything beyond smooth (and agile) filming work. I dream of putting a pixhawk (or pix2) in one.. hopefully one day someone will pay me to do that!
Aeronavics, not a big fan of their older frames, overpriced and poorly designed if im honest. The newer generation stuff and what they are currently playing with is very nice though, but has a price. They also now have some talented people working for them (friends haha). Ive retrofitted 5+ Older aeronavics/droidworx frames with pixhawks now, as wookongs,nazas, and A2 are about as useful as potatoes for serious work.
People pay silly money for commercial/industrial multirotors... because thats what the market seems to support!
If you want something fancy out of the box that may not totally fit your job, plenty of nice options out there (stuff like the Sensefly Exom for inspection work looks very impressive), If you want something built specifically for your needs though hit up some of the pro builders out there for a machine built for you, the problem is that there are a lot of hackjobs building crap and calling themselves pros, distinguishing between the pros and hacks isnt easy... some of the hacks have nice websites :D
What a great thread. The "higher end" drone companies are in a tough spot actually. They have to jump through all the regulatory hoops (battery certifications are particularly nasty) and any changes they make require them to recertify. Mistakes become very costly. Imagine trying to nail a system this complex without making any mistakes. And they have to make everything: vehicle, battery system, ground station, charger, radios, camera(s), gimbals. It isn't a product it is a very tight integration of multiple products. All the tooling that is required, suppliers going under, undocumented component changes, parts going end of life. It is a real nightmare. The volumes they do require tooling but don't really support a lower price structure in most cases. Add to the fact that they sell to military and government agencies (long expensive sales cycles) but they legally can't sell for a lower price to somebody else. Now it looks simple to take an internet recipe and build yourself an industrial machine. But try weatherproofing it, and making it packable, wind capable, and easy and safe enough for non experts to use. Even the best people here would have some trouble with that short list. I've seen what complexity non experts can manage - Mission planner is way too hard for them.
Yes a highly reliable/effective system could be built from parts for less than $10,000. That doesn't mean it is the the best thing for your business to do. Your business should leverage your expertise/contacts/interest with the best possible outcome for your customers and the lowest possible risk. It will take you quite a bit of time and effort to get to the point where you could build a competent industrial system. Before you do you will have many crashes when you thought you were ready. That is just the nature of complex systems.
I would say you need to get a phantom 4 and play with it. Use it to try to do the applications you envision (even on fake inspection targets) understand your application as best as you can and write down everything you can think of. Do lots of homework (this thread is brilliant). Make sure your customers will tolerate some incidents because there is a decent chance they will happen. They should know/understand the possible risks. Power companies lose full helicopters occasionally so they know mishaps occur. After that approach someone knowledgeable who can help you build/supply a system that suits your needs. You won't get everything on your list but you have to have that list as a starting point.
In picking a system/supplier beware of the lone case study. I've seen inspection pictures that were amazingly clear with a resolution that allowed zooming in to read small serial numbers from 15 m. Trouble was that was one of 22 pictures that were taken. The rest were all blurry. It is easy to make that system look good by cherry picking the photo but it wasn't a practical inspection system. You need efficient repeat-ability in the required environmental conditions. The best sign is a reference customer the company has that has bought many system over a period of time. It means they are successful with that system. Too many companies think the solution is simply controlling a vehicle with a high end camera on board. It really is more complex than that.
My opinion is won't save any money building a system on your own and you will more than likely loose customers/your whole business very early on if you try.
This exact question is what led me to start FreeBird Flight. I felt that their was a big disconnect between pricing and stated performance specs. I also felt that existing units were not safe enough for a lot of commercial use with large free-spinning rotors and limited ability to operate outside of benign weather conditions. We have developed 2 products over the past couple of years: (1) Our SurroundFrame, which is a unique proprietary carbon fiber frame that structurally encloses the blades and provides a number of other unique features; (2) FreeBird One: our high-performance sUAS that combines our SurroundFrame with an optimal set of components (including Pixhawk) to offer a high performance system that is completely weatherproof - heavy rain, snow, strong winds.
Our goal is safety and all-weather use without compromising performance. Flight time is still 30 mins, speed is 50-60 MPH on the standard setup, and payload capacity is 15 pounds-20 pounds.
We're also doing something else unique. We're using 3D printing for production. As a result, we can make design changes very quickly as necessary, either to improve the standard product or to implement a custom change requested by a customer.
We launched a Kickstarter last week to get some units into the market for feedback so we can make improvements and to get the word out about what we are doing. I believe the early bird units are gone, but the rest of the units are being offered for $2,500. Once we launch commercially on our website (after Kickstarter commitments are fulfilled), pricing will probably be $4,000-$5,000 depending on options, though we haven't set that in stone yet. Regardless, our plan is to come in considerably below the $10,000+ price that comparable models are.
I don't want to turn this into an ad (apologies if it already sounded like one), but this question hit home to me because it was the reason I started this business.
You can read and see more on our website: FreeBird Flight Website
Key factors in commercial drones.
It is not that hard to develop a flight controller for an aerospace engineer or 2.
FPGA= Fast response.
High quality radios= expensive
High quality components= more reliable
No frills= Get the job done
Ground station where required and support equipment= Package deal
Customer service and support= priceless
This adds up to about 30K minimum IMO
If you dumped all the features you don't need on the Pixhawk and put it in FPGA and take full advantage of high quality components it's about the same thing.
The code for the Pixhawk is bloated and in-flux all the time. Which is very good for hobbyist and very bad for commercial use.
If your in the business of commercial drones you have to have control of your code or your an idiot, fly by night or smoke and mirrors company.
If someone else can make changes to your code and effect your product your unreliable. Offering use of other code as "at your own risk" is fine. But your code has to be rock solid and you have to have a log system that allows you to control your code on your craft. You can't back up someone else s software. No way.
Ummm, whats wrong with a couple of Deep cycle lead acid batteries (couple of 2 or 3 x 100A/H batteries) in the truck with on board charger from the alternator. Drop the cost of the generator, no power supplies either so just saved the generator cost plus the power supplies and it's portable in that most utilities have service tracks that are close by.
Cant reduce the charge time though, so for continuous ops a few sets of batteries would be required.
@Monroe, That's absolutely true! It is really all about software. Hardware can be copied, but there is a considerable amount of R&D engineering involved in developing a controller board. Unfortunately, after a company spends that money, the hardware can be simply cloned. However, software can be copyrighted. It is very easy for developers to embed watermarks into code; used to catch copyright violators. A proprietary version of flight control software would take about a year to develop and cost about $500,000 to develop to a bug free state and fully functional state. Use of open source is probably why some UAV companies are failing and they are pulling out. I suspect many are in the process of developing a clean sheet closed version flight control software. As an example look at the SmartAP 3.0 flight controller; closed hardware and software.
Real pro's would use an ASIC if they could to make it even harder to copy. But I think in today's world using FPGA would be the way to go.
Depends on how many you think you can sell and/or who your selling them too.
You can work on and offer changes in your own code and you can offer (for a price) special features that can be developed for a certain customer. (which you might add to your product later) But YOU get paid to do that.
Don't let all the hype fool you some real companies will rise and others will fall as we go forward.
The ones that do as I said will be those that rise if they manage well.