Holidays Greetings!
There is much interest in seeing demo of our $50 pre-order NS-HP RTK receiver on a quadcopter, so we tested last week on a less windy day (with up to 4.5mph wind). A compass module and 5Hz version of NS-HP RTK receiver replaced the original GPS and compass inside the 3DR Iris. These two videos show the result we obtained with default values without any tuning.
Comments
@ Pascal P. Unlike normal GPS that gets position fix in seconds that everyone is accustomed to, lower cost single-frequency RTK receiver takes minutes to get RTK fix, longer it takes with lesser number of satellites. Dual-frequency receiver takes less than a minute. The experience of using single-frequency RTK will be quite different for anyone who has not used such product before. Thus we intentionally don’t offer such appear to be “plug and play” kits to avoid misleading customers, also quite frank in our literature telling customers with a bit of exaggeration that RTK fix time could vary, 1 ~ 10 minutes with GPS/Beidou RTK if in Asia; 1 ~ 20 minutes with GPS RTK depending on overhead satellites available, unlike others leaving out this important information. Shipping from Taiwan, radio transmitters has import declaration complications for customer, users could easily order a pair of radios locally, thus we intentionally don’t offer radios.
@ Nikola Rabchevsky Compared to Trimble R1, the DIY version in this blog, http://www.navsparkforum.com/viewtopic.php?f=37&t=306, costs 1/30 of R1's list price to assemble. Having more satellites to use, R1 will have faster RTK fix time while NS-HP will not get RTK fix in case overhead GPS satellite number falls down to 5 at some rare hours in some region. At other times when there are more GPS satellites overhead, NS-HP just get RTK fix a bit slower than R1 that use GPS + GLONASS. In Asia, if using our GPS/Beidou version, S2525F8-BD-RTK, that takes advantage of abundance of Beidou satellites (8 ~ 12 vs GLONASS 5 ~ 9), our single-frequency RTK performance is second to no one.
We began shipping NS-HP-5 for NS-HP pre-order customers starting last Friday, all will be shipped out by end of this week. Should be hearing user assessments on NS-HP-5's performance in some weeks.
Forgive me for asking the obvious but does this compare a much more expensive RTK system such as the Reach or a Trimble product?
@ Oliver Huang : Great product, very sexy for mapping application. You should propose full "ready to plug" kits on your website, including a pair of preconfigured base/rover receiver plus pair of radio.
"We hope to be able to offer superior performance multi-frequency RTK receiver at affordable sub-$100 price." : that would be a fantastic product, be prepared for mass production if you succeed to reach such a target !
Our drone pilot is not that familiar with ArduCopter code, so no modification done with firmware. Otherwise we would like to do better demo of stable hover using RTK altitude replacing barometric altitude, and tune PID for lesser horizontal drift.
We found default ArduCopter setting using 5Hz GPS, lower update rate result in some jerky movements, so 5Hz is needed.
Precision time/position stamp feature won’t be useful unless doing aerial survey photogrammetry. RTKLIB + raw measurement receivers cannot provide similar accuracy.
We hope to be able to offer superior performance multi-frequency RTK receiver at affordable sub-$100 price. Not possible using high unit price FPGA, only achievable with custom designed ASIC to bring down production cost; which itself costs $1M ~ $2M to develop. Pricing product differently according to features, like software companies do, makes lower price range products more accessible, and different profit streams allow faster self-generation of funding needed to develop next-generation product.
@Oliver; Thank you for detail setup information. Some other question on you setup.
Do you have any modification on ArduCopter3.2.1 software ?
One other question is your product variant of NS-HP series.
Which one is suitable for drone application ? And what is the reason of big difference of the cost ?
In our testing, a pair of 915MHz radio is used to send base station data to the rover NS-HP-5 on the quadcopter. Additional pair of 433MHz radio is used to send NMEA output in parallel to a laptop on the ground to monitor RTK float/fix behavior; this is not necessary for normal usage.
A ground plane is used to boost active antenna signal level couple dB more on the quadcopter; aluminum material is chosen to avoid interfering with the compass. A strong magnet inside the active antenna is also removed to avoid interfering with the compass. The quadcopter compass and accelerometer need to be calibrated with Mission Planner in order to get good result.
Our base station setup consists of an older NS-HP equivalent prototype board configured as base. The survey type antenna is mounted on roof of a car. Although a survey type antenna is used, lower cost active antenna should also work. User guide on NS-HP product page has details on how to configure NS-HP for base mode: http://navspark.mybigcommerce.com/ns-hp-rtk-capable-gps-receiver/
Typical base/rover usage scenario is as below figure shows, only additional antenna and radio is needed. The LED is used if wanting to know RTK ratio value before reaching RTK fix state.
They said RTK, then base RTK station should be there.
Looks great experiments, but very few setup information brought misunderstanding.
@NavSpark; Please elaborate as much your information on your testing.
That's really excellent ! Pls share link to your store.
Same question as Randy's here. Do you need a set of receivers?
Is does look really good.
Is there any additional hardware on the ground or is it just the GPS in the vehicle itself that was replaced?
-
1
-
2
-
3
-
4
of 4 Next