"Currently, this type of precise GPS is used mostly by surveyorsÛÓat a high price tag. ThatÛªs where Humphreys and his students come into the picture. They are working on making this type of precise GPS for a cheaper price ($50 instead of $5,000, the current going rate for centimeter-accurate GPS receivers), therefore making it more affordable for the mass market. The research is being funded by Samsung who hopes to build precise positioning into cars and smartphones."
gschrock, post: 389387, member: 556 wrote: This might not be as far away or far-fetched as we might think. Software-based receivers have been talked about for more than a decade
I am licensed as an amateur radio operator (HAM) and the big thing in amateur equipment is Software Defined Radio (SDR), several already on the market, so let's see, a GNSS receiver is a radio receiver, so yes it can and will be done and is already being done with other radios.
SHG
This whole concept of precise and accurate positioning of vehicles will be the driving (see what I did there?) force behind autonomous and intelligent vehicles allowing denser traffic on the same amount of lane miles, requires two things, knowing where the pavement is and where the vehicle is in relation to that pavement along with I suppose a few other sensors and a communications link.
SHG
gschrock, post: 389387, member: 556 wrote: This might not be as far away or far-fetched as we might think. Software-based receivers have been talked about for more than a decade, prototyped (often as rather bulky devices) and the mantra is always "soon". But have heard Todd present about the work of his team and it would not surprise me to hear that they've actually pulled this off . Hope to learn more at the upcoming ION+GNSS in Portland in a few weeks....
Would the antenna size and design be a limiting factor?
Also, in a car, with a phone or third party nav unit GPS receiver/antenna almost always under the roof, it would be using stray signals, I would think. Instead of rejecting multipath, the phones must rely on them sometimes, I would think. The concept of centimeter level accuracy would seem to mandate rejecting mulitpath signals, but what would remain inside a car at that point?
Now an antenna on the roof picking up 3 or 4 constellations, with what amounts to a huge ground plane under it, that could work very well, I would think.
Y'all mean I'm supposed to get into a car with a GPS driving, and not be nervous? I don't trust them that much! I can reshoot a point but I can only go off a bridge once.
I always find it amazing when the developers talk about the cost of things as $50 or less for the parts and when it becomes available, the going price is 1,000 times that.
A Harris, post: 389427, member: 81 wrote: I always find it amazing when the developers talk about the cost of things as $50 or less for the parts and when it becomes available, the going price is 1,000 times that.
That is the parts cost, but remember the research and development, investment costs (like right now they aren't making any real money on it, so big risk of no return), opportunity costs, etc. $1000 might be a steal.
I have a vague recollection about the cost of dual-frequency GNSS being partly attributable to licensing fees associated with some aspect of the hardware or firmware. Is that still an issue, or was it ever?
Just trying to understand. So I buy an after-market GNSS receiver providing me an autonomous (?) position of the antenna at the few centimeter level. Won't I need to precisely survey the relative locations of the ends and sides of the vehicle (at least) to make such accuracy useful?
Years ago, early 2000s, I used an Applanix POS-LV system to do some 3D mapping. Nifty system incorporating GPS, inertial sensors and a device counting wheel revolutions. Given it's proprietary software it was hard for me to be comfortable with the results. Careful attention had to be given to establishing relative locations of sensor systems.
Hard to imagine that the costs could be kept as low as advertised.
I can see this autonomous driving technology paired with something akin to Apple's Siri artificial intelligence system. An open road 'cruise control/navigation system. I can't see how it would work well driving through tunnels or under overpasses or in a dense built up down town area where GPS signals would be bouncing around all over the place or altogether absent. I'd name the system 'Jesus' so I could freak my passengers out when I throw up my arms and shout out 'Take the wheel Jesus!!!', and he does. :innocent:
No offense to anyone I hope. Have a great weekend.
Just because I'm paranoid, doesn't mean they aren't out to get me.
The Tech moves fast and the prices fall just like it always has: I bought a Piksi RTK centimeter set up (L1 only, it's slow) at $1k to play with then in less than a year ublox came out with the same thing L1,RTK with a rinex twist and way better firmware, at only $500 and I tried it for awhile (good system actually) now L1, L2 is supposed to be right around the corner at the same prices. Software is a big issue with all the new upstarts rtklib is good for what it is (free) but you get what you pay for. All these try to get to the levels you guys had with gps only. And now software based on top of cheap chips watch out.