http://www.wired.com/2016/03/dji-phantom-4-smarter-faster-super-sleek/
I don't want to like DJI for the same reason I don't want to like Apple - I like open systems that can be tinkered with and that a community can work on together to improve in the long term. But I'm more and more impressed with what they come up with and how they package it.
With all the recent changes between 3dr and APM, can either or both projects regroup to take them on and provide a compelling price/performance alternative? The solo is great but was too expensive to compete directly against the phantom 3, perhaps this more expensive phantom might actually be a good thing for the solo? Has DJI already won the consumer market for good?
Comments
I think "Snapdragon Flight" is nearest goal for ArduCopter. It has quantified flight controller run with PX4 or ArduCopter that is connected to Snapdragon 801. They have 4K cam, stereo camera and flow sensor.
@Maxime
Yes, they have a surround SDK based on nVidia Jetson TJ1, developed by fudan Uni. now stereo on P4
DJI got SLAM? Do not think so. Photogrammetry?
Otherwise great input!
DJI has SDR, SLAM, FOC motor control, good video codecs and now has integrated into a good priced package.
Now building a quad too shoot your weekend party is not as interesting when 5 years ago, but the market for small business building custom UAV solution for contract demand is rising.
For RC model builder now are having fun with Nase32, they don't need GPS or UFO like steady hovering, and many of them i know don't have the patient to build something like ardupilot skill sets requires.
where are we going now on this?
I will choose a open SoC platform like nVidia Jetson to begin, with better MEMS gyro, and a ubut 802.11 dongle. But the price to play with these toys will be high. forget your RPi, build a form factor and mechanical reliable super computer on a >10lbs platform, get people from gnuradio, ffmpeg, ROS, and many small companies in China, build a open source reference design for ROBOTIC areo computing platform.
No linux skills? better get some, if you want to make a drone with these fancy funtions. if not, fly your old style MCU APM, but later PX4 may have better application.
Hello, this is kind of my first post on the forum and, although the topic was on P4, my point may be suited here regarding the direction the talks went.
I tried to raise some issues I will develop here on RCgroups http://www.rcgroups.com/forums/showthread.php?t=2602644.
As I have learnt about C. Anderson talks, the future of drones is tied to Linux boards (https://www.youtube.com/watch?v=1DqVl9P1FcY). May we say real-time Linux distribs + lasers + modularity + integration? All about those.
The FPV and aerial filming consumer markets are mature, why target them? Craig Issod is right, what is the goal here? We are not 3DR, best of luck for them. I wish the goal was enforced as "staying cutting-edge".
P4 as stated did not revolutionize anything about reactive autonomy such as obstacle avoidance, but brought it brillantly to the consumers. As Scott8 said, there is an urge for the open-source community to catch-up with great real-time obstacle avoidance fast.
As Matt said, the community should not have a inferiority complex about a billionaire and his engineers, but neither should despise "dumbing it down". Integration and packaging/refinement are utmost needed for any hobbyist or entry-level consumer.
Muscate said (no quote! :0) "What I'm missing to go from a pixhawk to a next gen platform though is an integration between the flight controller, video processing and transmission capabilities" How can we not have DIGITAL h265 video feeds going to the board for both storage and LTE transmission purposes? It has been years since the lauch of Lightbridge. How can we still need OSD, telemetry devices, when all processing could be done on the board? How can all the radio transmission not be using one single full-duplex-channel antenna?
But all the stakes are in cognitive autonomy, I take the term from this good Nature article : http://www.nature.com/nature/journal/v521/n7553/pdf/nature14542.pdf. We need more integration (the flight controller - companion computer architecture is great), software-defined radio, swarming, SLAM.
Phased-array lasers are a great opportunity for newcomers from 2017 and beyond http://techcrunch.com/2016/01/08/quanergy-s3/. Other great techs like VCSELs will go down from the academic skies. Full 3D SLAM based on laser and reactive autonomy combining laser and optic flow systems are already done by researchers and reachable for the community within years. There is also just so much work to do with swarms. Will Dronecode be part of it. I personally do not care if it happens with Ardupilot or Paparazzi or a continued open-source PX4. I just hope so much, it will not be only DJI, Parrot or closed 3DR : will not give any money to any Bill Gates wanabee of the drones, no matter how great SDKs evolved.
A lot of time seems to have been wasted with Ardupilot with Solo. This would not be the case if Pix2 was out there and and we had a better SDK to use with it. As it stands there is an effort to get it out but it seems not many willing parties to commit the money to bring it to market (I can stand corrected but the GPL designs were posted some time back).
It is a good question though. Can we get obstacle avoidance for Arducopter? Is it a feasible DIY thing or are we going to need integrated payload modules and more extensive and better documented communication protocols between FC and Co-Processor. And is anyone going to make this hardware and support the required firmware dev under a GPL license? I am not sure I could make that case to an investor. Seems like a Navio-like company could do this. I know that is not a popular sentiment with some people.
Ben, now you are in my wheelhouse!
But I have to respectfully disagree with your summary.
This industry has never - and I mean never - seen a marketing and PR push like the 3DR Solo. Frontman Colin Guinn and marketing leading Oren are really really aggressive and spent many millions of dollars on marketing. You may be looking at the wrong 3DR FB page
"https://www.facebook.com/3drobotics/?fref=ts"
160K likes
Do a google search on 3DR Solo and you will find thousands of marketing pieces (PR) repeated by most every major media. You name it - Forbes, Fox Business, Verge, CNET, Popular Mechanics and thousands of other sites - all printing very positive pieces before the Solo ever existed.
The story of Solo is certainly not one of bad marketing. It is more one of premature ej....well, you know what I mean. The Marketing and Sales and everything else went into full swing 6 months before the Solo was starting to be up to their initial claims (gimbal, GoPro control, etc.).
By that time it was too late. Now they have the completed units but no money left for marketing. In fact, one buyer was complaining yesterday that they couldn't get the site to work so they tried to call to give their CC number. Well - as you may know, 3DR stopped taking phonecalls.
But as to their marketing - if looked at it independently of the product (quality, honesty, etc.) and of budget or ROI, it was very slick.
Think about it - if it wasn't how did they get BestBuy, B&H and all the others to sign up?
Take another look at their real FB page (there are fakes). They also have 70K or so twitter followers. Threads which talk about the Solo have as many as a million views.
It was the lack of a finished and polished product at a competitive price which caused the slow sales (IMHO).
Well I'm surprized nobody mentionned the marketing about the two brands. I know that the diydrones community is not marketing oriented (which is probably a good thing), but it's certainly an aspect that can't be neglected.
Maybe dji has a larger budget for this, but a simple google search shows 20 more results for 'dji phantom' than for '3DR solo', same for the news search.
And to keep it short, the websites of the two companies are very different, and I think dji's site is more attractive, even simply by its white color scheme (probably inspired by Apple).
In comparison 3DR's black and green scheme seems less attractive.
And OK just to finish, on dji's shop there are lots of promotions. It may seem not important, but we are emotional beings and are certainly attracted by a large choice of products and "-35%" labels.
Nothing similar on 3DR shop, and a very limited choice of products :
"solo with preinstalled gimbal + backpack"
"solo with gimbal"
"solo"
"solo gimbal"
Solo page : only 83 likes on facebook, only 9 "+1" on Google, something is wrong with its communication, even if the product is (probably) excellent.
I wish we had rudimentary obstacle avoidance and optical loiter, etc, in APM. It's been worked on by various people for years, but the official response is always that it sort of works, but not well enough to include it in any releases. Well if DJI are releasing it, and people are saying it probably doesn't work well... Shouldn't an open source experimental platform like APM have it available.. even if it doesn't work well?
I worked w/ robotics and sensors, etc, in college back in the mid 90's, but my career path took a different direction. Then when I got in to drones a couple of years ago, I was shocked to see that pretty much nothing changed from when I was playing with this stuff 20 years ago! Everything else is 100x better, but detection of a large flat object (automatically avoiding walls or the floor) still seems beyond our reach?!?
the solo got his chance.. but the biggest misstake was the goto-market.. the hardware was not available in time and here in europe the solo is still not available. many people here whats buy this solo but they can't so they buy the phantom 3...
now, the solo and 3dr lost much Reputation in europe.
-
1
-
2
-
3
-
4
of 4 Next