One of our employees who drives our GPS Insight Scion in a car pool took it Saturday morning to a remote area of the desert to go for a run.

He parked off the road, on a pull-out which was nowhere near where cars drive.

He came back to this $90 parking ticket (with my name on it since it’s my car officially…):

$90 ridiculous parking ticket

So here’s how you prove that Maricopa County Sheriff’s are scrounging around for money for our broke County, using GPS Insight:

Quickly run a 3D vehicle history for the Scion for that day

Open the “Time Slider” to the first trip of the day & show just the dots (green) on the map corresponding to the drive to the first stop and 45 minute stop (yellow):

History shows the drive to go to an early AM run

Then Zoom down & see the vehicle was parked off the road itself:

Parking off the road yields a $90 ticket

Then go to Street View mode & see that it’s a paved pull-off.  Why wouldn’t it be ok to park there to go for a run/ride/hike?

Did they pave it just to lure people in to give them $90 parking tickets?  Thanks Deputy Thompson.

Street View shows this is nowhere near the road

I’ve paid it since it’s got my name on it, but I will definitely send a copy of the blog article to the Judge & see if we can get it refunded.

How ridiculous, though.

Even though politics and a County with financial issues trumps GPS evidence sometimes, at least you can clear the suspicion that your driver did the wrong thing.

Also, using our DriverID, you know which driver was in the vehicle at the time of the ticket so you know who to talk with, if the original got “lost.”

Rob.

My wife loves the fact that she’s a guinea pig and has a GPS tracking device in her car… (not really)

But it comes in handy really often to know where she & the kids are.  This article shows how quick and convenient it is to use GPS Insight to answer real questions in a useful way, daily.

I was heading out of the office, waiting at the light to head to the store, and a car that looked like hers drove by toward our home.  All the cars in Scottsdale look the same and all the windows are tinted, and we don’t have a vanity license plate I can remember (which probably is the way to know it’s my car…).  I wasn’t sure if it was her heading home from somewhere or just one of the other thousand SUV’s in town.

I thought maybe she would want to turn around and grab dinner with the kids where I was going.

But I didn’t know if it was her car or not, so I pulled out my iPhone & checked the map of all our vehicles.

Within seconds, I knew it was her, thanks to the GPS Insight fleet tracking product.

Here’s how I did it in 10 seconds:

I simply refreshed the link for GPS Insight’s “Mobile Maps” then “zoomed down” to my location using the “zoom to me” button (I don’t know what it’s called, but I have an arrow pointing to it below).  Clicking on the vehicle just North of the intersection shows it’s my wife’s vehicle (Nav2):

Quick map tells me it was my wife (Nav2) who just drove by

Then just to show I’m really at the light waiting to turn left, I click on the pin there (I have 2 devices in my car so both are me):

Me at the light waiting to turn left

Then I can show where the cars & the office are using Google Maps’ street view (by pressing the little orange guy shown above):

Google mobile Maps shows the intersection where I'm sitting, etc.

This isn’t life & death or even business, but gives a good illustration of how fast you can answer questions about where people are at (or perhaps which of your techs just blew a red light in front of you a second ago).

10 seconds, and you’ve got your answer.

Rob.

I took my  boys (7 & 9) skiing (snowboarding they correct me every time…) up to Flagstaff AZ this weekend.

I brought my Garmin Edge 705 (a Garmin for bikes, basically) & put it in my coat pocket to get a really good feel for where we went during the weekend.  It logs every 10 seconds for eventual upload (vs. once a minute in real time for the EZ-1000 I brought along as well).

Garmin EDGE 705

They are really nice, and we don’t sell them — I bought mine at the bike shop.  [We do sell Garmin's for vehicles and can integrate them with our GPSI-4000 GPS tracking solution though...]

The Garmin is an optimist, and thinks I ran up all those hills & burned 5248 calories in the process (had I brought the wireless heart monitor it would have known better):

Garmin's interpretation of my Skiing with my kids

This is a great image of GPS Insight vs. Garmin (bear in mind we are an “ACTIVE” tracking system whereas the Garmin is “PASSIVE” and needs you to upload the data eventually):

GPS Insight vs. Garmin for ski tracking

The “pins” are EZ-1000 points [every minute while in cell coverage, which is spotty on the mountain].  The yellow circles are Garmin points (a LOT more of them).

Here is the Garmin unit track of where we went (and where I remembered to turn the unit on…):

Skiing at Snowbowl in Flagstaff AZ

It’s nice to have that for sports usage.  But if you want to track your vehicles in real time, that’s not an option — you need an active tracking solution like GPS Insight. Both are great products — just for entirely different needs.

And here are my cold kids.  Lots of snow this weekend, and way colder than they’re used to living in the desert.

Rob's kids in their cold-weather snowboarding attire

Rob.

I pretty much have a never-ending supply of EZ-1000’s so I brought one skiing with my 2 boys on a recent trip.

Jack, my older son, is old enough to go skiing (boarding, he would correct me) without me. Actually, he has officially passed me by — he does black runs I refuse to do (small on a board is way better than big on skis when it comes to moguls).

I wanted to know where he was at one point so I ran a quick mobile map on my iPhone & put it in “compass mode” so I could see what direction he was from me.

Here it is (I’m the blue dot, Jack is the red pin):

GPS Insight EZ-1000 tracks my snowboarding son on the mountain

Here he really was (my eyes are better than the 3 megapixel camera on the iPhone…) — note that he’s between the lift & the ski patrol “house” just like the map shows it:

GPS Insight EZ-1000 tracks my snowboarding son on the mountain

I ran a 3D history of that device for the 2 days I remembered to bring it and put it in Jack’s pocket & it puts him exactly where my iPhone shows him at 1:44 (note the time in the first screen shot). Waiting 4 minutes at the bottom of the hill for his 2 friends:

Showing skiing activity using an EZ-1000 from GPS Insight

Also interesting is the straight lines which depict the lifts very clearly. The main lifts are in the “clutter” of dots on the left side, but the lifts we went on once each are really easy to spot toward the top right.

It’s easy to see which runs got the most use by turning off the “time slider” and looking at just the blue path:

GPS Tracking my son on the ski (board) slopes

Here are my two boarders:

Jack & Ryan on a snowboarding trip

And by looking at the GPS track as well as how well he was jumping and grinding, I can tell Jack went through the terrain park most of all:

Jack grinding/jumping off a box in the terrain park

I’ll try embedding a Facebook video I have of him going through the terrain park here:

Rob.

My 4 year old daughter Sarah got a Disney Princess jeep from Santa Claus on Christmas (along with a liberally applied Tinkerbell makeup kit, earrings and much much more…).

GPS Insight tracks Princess Sarah

Never missing a chance to track something important, I covertly put “Rob 1000,” a GPS Insight EZ-1000 in the battery compartment:

GPS Insight tracks Princess Sarah

So off, we go. When we get to the corner, I check the location on my iPhone & it shows we’re there:

(I’ve blurred the street names):

Tracking Princess Sarah

Here’s a picture of her at that intersection:

Tracking Princess Sarah on Christmas

A quick 3D history report:

GPS Insight 3D History GPS Tracking map

… and here’s our 20 minute walk around the neighborhood. Per usual, our 3D mapping starts “light & thin” and gets “darker & thicker” so you can see the direction of travel is clockwise around the neighborhood easily, just by looking at it (or by the times which are at 1 minute intervals): [by the way, blue = idling, e.g. zero MPH, & green = moving]

Princess Procession on Christmas Day

She wants a faster car though already…

Princess Sarah wants a faster ride

If there’s one person I’ll break the “we don’t track kids” rule at GPS Insight for, it’ll be Sarah when she’s ready to drive (especially at the rate she’s applying her Tinkerbell makeup).

Ok, here’s one more bonus picture of my Princess:

Sarah on Christmas

Rob.

We will have a formal announcement about this after the holidays.

Just last Tuesday, we lit up exactly 1,000 new units on GPS Insight for a new customer.

They all came online at the exact same time, which is not typical, but this had an interesting and noticeable effect on our daily processing.

Here is a graph we get each day which is typically 100% bell curve shaped. This indicates driving activity, and peaks during the middle of the day when most of the drivers of our tracked delivery vehicles, service vehicles, and government vehicles are out doing their jobs.

GPS Insight turns on 1000 new units

It was very evident when these devices started reporting to GPS Insight, so I thought I would point out the nice “bump” we got to our overall vehicle installed base Tuesday.

It is interesting to note that even though we have vehicles in 4 different timezones (6 if you include Alaska and Hawaii), and in 100 different types of business, they all wind up “smoothing each other out” to a single bell curve.

The street sweepers and a number of over the road/long haul vehicles work at night typically, which keeps our nighttime activity from dipping too low, and the early morning service workers (Construction, typically) get things off in a hurry starting around 4 AM MST (here in AZ this time of year that means 6 AM New York Time). The longer tail at the end of the day is because of overtime — drivers get going according to a schedule, but don’t always finish on time.

Some drivers drive to a workplace once a day, then there is no more movement until they leave to go home, and some drive all day long (e.g. delivery vans). When you you put together tens of thousands of vehicles though, across over 1,000 customers, things balance out and become pretty predictable.

GPS Tracking histogram / Bell Curve

Here’s a daily “by hour” for the whole month. The only anomaly is a slight dip in the 11:00 hour — I’m pretty sure that’s lunch related. I bet if our customers check the street view on their vehicles locations, there would be a lot of this stuff going on:

This is our Scion 4000 on the way to take a couple salespeople to the airport to head home after a week in the office:

19 minute lunch stop into the 11:00 Hour

In-N-Out stop for Joe Vidmar

Our Chicago guys only get into Scottsdale occasionally, and needed their In-N-Out Burger fix before heading back on Friday. Them & several other thousand drivers being tracked by GPS Insight around lunchtime. And now 1,000 more.

Rob.

So I have 2 back to back soccer games today and while the kids are off warming up, I thought I would figure out how to write a blog post from my iPhone using a new app I downloaded.

I have to keep if about GPS so here is a screenshot of my location under the tree and my car’s location in red with the length it’s been stopped.

Jack just stopped 4 shots on goal. I should probably watch the game now…

Afterward, they had the obligatory pizza party:

Soccer Pizza Party

And how much time did we spend there, and at the games prior? Just run a quick (1.8 second) stop report for the day.

GPS Tracking stop report for soccer activity

Easy to see — 3 hours, 9 minutes at the games, 1 hour, 3 minutes at the party, and 17 minutes prior to the games getting snacks for the game at Target

By the way, the iPhone app for the blog is not very good. I had to use a browser to move the pictures around, and the sizes aren’t ideal. Oh well, I should probably be watching the game instead anyway.

Rob

We were lucky enough to get a couple passes for this weekend’s NASCAR event, to include pit passes to the garage & pits.

We had a couple of EZ-1000’s with us, and I want to show how easy it is to quantify how much time we actually spent in the pits (vs. the stands/box):

I first create a landmark called “PIR Pits” around the pits:

tracking activity in the NASCAR pits

Then cut & paste the PIR Pits landmark into GPS Insight:

Paste a geofence into GPS Insight

Paste here:

Paste geofences into GPS Insight

Confirm here:

Paste geofences into GPS Insight

Run a report here:

GPS Track of Pit activity at NASCAR using GPS Insight EZ-1000's

1.9 hours in the pits

Except that missed the part where we went really close to the “inside the building” chatter from the GPS device, so I had to carefully re-draw the polygon to get a more accurate picture of the true time spent down there.

So even though we may pick up a little bit of false “in the pit” activity since it’s so close to the box/grandstand (with drift due to the device being enclosed), this adjustment to the “pits” will give a better representation of how much time was there:

Extend the boundary of the pits geofence to get all activity

While we get a couple of random “1 minute” visits which are inaccurate due to the EZ-1000’s “inside” drift, the 2.8 hours is much more accurate than the original 1.8 — both Brent and I spent half an hour in that remote area of the pits and it makes a big difference to the total amount of time in the report by accurately creating the polygon:

2.8 hours in the pits

A fun (and loud) time was had by all:

NASCAR in Phoenix

And a favorite of mine, the RedBull 83 car:

RedBull NASCAR

I like this car so much I bought the $60 model for my bookshelf. Brian Vickers may have come in 38th today, but we drink a lot of RedBull and don’t have much use for Lowe’s in our office.

Rob.

I decided to take my 9 year old, Jack, to Camelback mountain yesterday. I was curious how much longer it would take than the last time I went.

I brought an EZ-1000 & here is a picture of our hike, which was 1:30 up, & :56 down:

Camelback Hike with a GPS Insight EZ-1000 Tracking device

Here is a picture of Jack at the bottom:

Camelback Hike with a GPS Insight EZ-1000 Tracking device

Here was our location, courtesy of the iPhone (blue dot) and the EZ-1000 which had reported just a few seconds earlier:

Camelback Hike with a GPS Insight EZ-1000 Tracking device

And 1:20 later, at the top:

Camelback Hike with a GPS Insight EZ-1000 Tracking device

With sweaty Dad:

Camelback Hike with a GPS Insight EZ-1000 Tracking device

And one of Jack’s shady cave — he found a few of them on the way up & down to rest in:

Camelback Hike with a GPS Insight EZ-1000 Tracking device

It took a little longer this time (last time 1 1/2 hours, this time 2 1/2 hours). But my heart didn’t feel like exploding as much as when I was in a hurry. Plus I had some company. Much better this way.

Rob.

I took my kids and wife to see Wicked (the musical) last weekend at Arizona State.

It was hot. 118 degrees hot.

The play was good, but my kids wanted to walk over this bridge to leave afterward, which left me confused on where I had parked my car. The heat was melting my brain.

So I pulled up the vehicle location on my iPhone using our Mobile Map application:

The blue dot is my iPhone & it told me that my car was East of me. That’s all I needed to know, but didn’t necessarily know what direction was East (Gammage is a big round spaceship looking building so direction is not immediately apparent).

The new iPhone has a compass built in — perfect. I wish I had this when I was in the Army:

So I head East & see my car and note that it’s in the 6th space:

Looking at the iPhone in satellite view & zooming in, we show I’m in the 5th space, as well as where I am when I pull this up out of curiosity:

So then we get going & even after 10 minutes of driving, it still shows as 118 degrees. Note that the AC & the seat coolers are cranked as high as they can go…

We were in a hurry since the play was really long and our babysitter needed to leave at 5. Unfortunately at one point I saw we were in Mesa and I realized I missed my turn, costing me 7.42 miles while I drove the wrong direction.

I was a little preoccupied trying not to burn myself on the steering wheel, etc. to notice I had passed up the 101.

Apart from it being insanely hot, and my missing my turn to get home to the babysitter on time, it was a great day — This is my second time seeing Wicked & it’s a fantastic play (The Wizard of Oz from the Wicked Witch’s perspective).

My 7 year old son Ryan had this to say: Wizard of Oz – 1 star. Wicked – 4 stars. He’s hard to please but Wicked worked better than a 70 year old movie apparently.

Rob.

I decided to hike Camelback Mountain today — I’ll try to make this short (shorter than the 1 1/2 hour hike up & down the toughest mountain I’ve done in the Valley).

I couldn’t remember where to go so I looked it up on the internet & then dispatched myself via my Garmin from my PC:

Then I created a landmark as well, while I was at it (note I put the “route me here” address where it should go, but then I outlined the whole mountain) :

Then I routed myself there with my Garmin by wirelessly dispatching my vehicle’s Garmin to that location:

Turns out it was the wrong entrance — I go to the other one typically, so I drove around aimlessly finding that, plus the REI sports place where I bought a new water backpack (called a Camelbak, coincidentally…) .

Running an efficiency report, I ran 58% more time and 20% more miles (since the miles I went extra were on surface streets, not highways, this makes sense):

Here is my driving around aimlessly, wasting time & fuel (green dots, vs. blue “optimal path”):

Then I get to Camelback, and send a couple SMS notes when I start, get to the top, then get back to the bottom:

Here is my track of the 1 1/2 hour climb to the top and back (using an EZ-1000 at 10 second updates):

Here is an iPhone picture of me at the top, mapping both my iPhone GPS location (blue) and the EZ-1000 location (red pin):

But here is the good part:

Toward the bottom:

And at the top:

I’m a little sore after the 1 1/2 hour long trek up & back down, but the view is worth it.

And I even wore a GPS Insight hat…

Rob.

So yesterday I took my 9 year old son, a friend and his son to the Diamondbacks game (they lost…). Thanks to Mike Greco at bluemedia for the tickets…

On the way home, after dealing with the 101 being closed (Arizona doesn’t know how to work on roads without closing them entirely), yours truly got pulled over 2 miles from home after punching it to make a stale yellow light. In my defense, there was no one around (except the police officer apparently, who I never saw).

Anyway, I got pulled over shortly thereafter and was informed I did 67 through the light in a 50. Oops.

I haven’t had a ticket for 15 years, so there goes that streak.

I got home shortly thereafter & guess what, he was right.

My GPSI-4000, which takes speed samples once a second, got me doing 68.

Here’s a picture which pretty much tells it all:

Here is my idle stop while receiving the ticket:

I pulled over near a community’s entrance to get over from traffic:

I brought an EZ-1000 with me for my son to hold on to at the game in case I lost him, & it was set for 10 second updates.

It got me at 67 MPH going through the light too (and shows I immediately slowed down):

So, the moral of the story is: I was speeding. It was literally for just a few seconds to catch a yellow light rather than slam on the brakes and wake the kids up (that’s my story), but both the police AND my GPSI-4000/EZ-1000 caught me. FYI, the GO-3000 is equally accurate, and our 3500 lineup checks speed every 20 seconds, so it catches speeding, just not the rapid up & down speeding like I exhibited yesterday — it got me at 62 MPH. To put it in perspective, many of our competeitors check speed once every 5 minutes and don’t report max speed — just instantaneous. We report max, instantaneous, and average, and this is detailed in several “speeding” related blog articles.

On a less depressing note, we had a nice time at the ballgame, and stopped at Alice Cooper’s Cooperstown before which we tracked on the EZ-1000. I landmarked it while I was there by sending a text message of: ‘gps rob 1000k landmark cooperstown’ so now it shows up here like this:

A quick landmark report shows we were there for 41 minutes:

41 much more worthwhile minutes than sitting on the side of the road waiting for a ticket 2 miles away from home.

I’ll let you know how traffic school works out.

Now that I think about it, had I been using my new Garmin routing capability I would have been directed to take the shorter path home after getting detoured, and would have avoided this ticket altogether… Grr…

Never mind all those other light green speeding dots where thankfully there weren’t any police or speed cameras.

Rob.

I live in a beautiful area with mountains EVERYWHERE.

Before I moved here, I would climb them every trip.

Now that I’ve lived here, I never do.

Today I took my 9 year old son and his friend to climb (hike) Pinnacle Peak, an easy one nearby my house.

Just checking the map of my trip for the day, and it’s 7.8 miles there — we stayed for 1 hour 23 minutes.

You can see that the parking lot was full so I parked outside on the street, and I got there at 8:40, then left at 10:04:

Here’s a picture from near the top:

If I want to know how recently I was there last, I can quickly create a landmark from a stop report by clicking here:

Then create it, assigning to the Arizona landmark group while I’m at it:

Going all the way back to February, running a landmark report, I find my last time going there:

Note that you can go back >>> 3 YEARS <<< with GPS Insight (or more if you ask nicely) to get information like this.

Most competitors only let you go back 90 days & charge a huge amount if you ever need the data beyond that from them.

The last time I was at Pinnacle Peak was back in February, on 2/2/09, for 1.4 hours.

I need to get there more often. Now that it’s hot again, I think I may need to wait again until Winter.

Rob.

GPS is very accurate, but still not 100%, especially for civilian/commercial purposes.

Typically, when a vehicle is moving, we get it within 5 feet of accuracy.

When a vehicle sits still for some time, tracking devices (all of them) typically experience “GPS Drift” of up to 25 feet or so.

However, GPS Insight fixes and insulates our customers from this inaccuracy and I will show you how in this article:

This is what GPS Drift it looks like using a version of our mapping which only our support staff has access to — this particular map shows how long each individual point takes to make it into our database, and also shows every single reported point, and I use it make this “GPS drift” evident:

These are roughly 24 points which bounce around up to 25 feet from the center (they are 30 second updates for 12 minutes while my wife & family waited for me to get a few things from the office on our way home tonight):

All of these points were in our database within 4 seconds of them being registered by the GPS device (in this case, a GPSI-4000 at 30 second updates).

However, we “consolidate” them into a single, highly accurate 12 minute idle stop, which shows precisely where we parked the vehicle and let it run for 12 minutes — the parking lot is practically empty and we parked in the 2nd space:

To show what our customers see, I run a typical 3D history map to compare:

And we only see one “pin” where my vehicle truly parked for 12 minutes:

Note the “lighter/thinner” red line from earlier in the day (where I drove through the ATM at the bank to our North earlier). We show activity in this way to help easily determine direction and time of travel with the “path” which exists below the individual location “pins” which we show.

Here I have zoomed out a bit, and we see this 2 minute idle stop (at the ATM drive through), 5 1/2 hours prior to coming back to that area (it’s a long drive to Tortilla Flat):

The light/thin to dark/thick lines are another way we help customers gain easy insight into their fleet activity.

Here is my vehicle (Rob 4000) and Brent’s — it is immediately apparent by looking at the line thickness/color what direction of travel I took to the office, then back home (by way of the post office) today. It is trivial to see that Brent drove in the direction of travel from light/thin to darker/thicker.

These are the types of things which you might not immediately notice with GPS Insight. Things just “work” the way they should. But behind the scenes, we are processing and filtering and consolidating and sanity checking millions of points per day for our customers to make sure they see sensible data and can use that to bring efficiency and insight to their businesses. We take advice on usability from customers all the time, and encourage them to provide feedback to help better the product.

I just happened to run a map tonight for a different reason (we were testing a different unit) and I noticed this idling activity. I thought it would be a good time to detail how we do these two (of many) convenient map enhancements for our customers.

By the way, here is an iPhone pic of where we were at with the kids (’Tortilla Flat’ — an old mining town/tourist trap in Apache Junction):

It is a beautiful area near the Superstition Mountains — we thought the drive would be nice on this Memorial Day weekend. I’ve had better food before though…

Note the hitching post where people tie their horses up when they can eat at the Saloon.

We don’t have GPS or diagnostics for horses, and note they left different types of “emissions.”

Rob.

I got an odd-hours alert for our GPS Insight Scion (the wrapped vehicle we use to do advertising, occasional installs, etc.).

One of our employees took it home tonight, which is fine — miles & miles of cheap advertising since he lives far away from the office.

Except they’re speeding so I looked online to see what’s going on. (I also got a speeding alert…)

The Scion has 3 units installed — a 3500, a 4000, and a 3000. The “Scion” vehicle is the 3500, at 2 minute updates, and the others are at 1 minute updates.

This was a pretty interesting distribution of whereabouts on the 60 heading West, all speeding, but nicely spaced out between location reports:

I just thought I would share, it looked interesting.

FYI, the “out of range” “Pedestrian” units are our EZ-1000’s — when you turn them off, to save on battery, they rightfully go “out of range” – since we only have a few for testing on our account, they are all turned off at night, since we’re not out patrolling the mall, etc…

A few minutes later, the 3 units in the Scion are still equally spread out — this time Alena is in chase. I’m guessing she’s trying to catch up to her husband, who is driving home from wherever they met for dinner together…

And here we go, the culprit is…. Grant, stopped finally at Grant House.

[The night after I wrote this article, 3 of our installers are driving the Scion to coordinate a large install in California together, and one has an EZ-1000 with him (don’t ask me why he named it ‘Ghost Rider’ — I have no idea). Now there are 4 different types of tracking device in the Scion, all together, again all speeding… That’s California though, they’re probably getting passed left & right.

Here’s a picture of the 4 units, off by just a bit time/reporting-wise, all in an alert mode. Bear in mind the speeds are slightly different due to them having different 1-2 minute sample timeframes:

Rob.

I forgot to turn off an EZ-1000 I had brought on a trip. It was in my overhead luggage (oops).

It still tracked me on the runway (a max speed of 214 until it lost signal), then all the way home while in my trunk.

Run a quick 3D map for Friday:

We took off on time (3:35 flight pushed back at exactly 3:35 & left the runway at 3:40):

Then the unit last reported before losing cell coverage at 214 MPH after turning over the ocean:

I was out of cell range for roughly an hour, and covered 330 miles between Orange County, CA and Phoenix, AZ which means we averaged about 330 MPH:

All of our units except for this one (the EZ-1000) would store that history, but the EZ-1000 is more about where a person is right now (e.g. security guards, police officers) so it does not store data if it loses cell coverage — since we rarely lose cell coverage (except in planes at 33,000 feet…) it’s a non-issue, as you’ll see next.

After landing, it picks right up again, then tracks my vehicle all the way home, even though I had the unit in my suitcase, in my closed metal trunk:

And on the freeway, it is exactly accurate (but at 2 minute updates) relative to my GPSI-4000 at 10 second updates in the same vehicle (red line=10 second with the GPSI-4000, blue line = 2 minutes with the EZ-1000):

Here on the highway there are 2 points 4 seconds apart, with the same exact speed (67 MPH):

The points are 56 feet apart, which seems reasonable for 4 seconds at 67 MPH (technically it should be 393′ but the 1000 takes a couple seconds to transmit vs. the 4000 which is pretty much instant — but close enough– we’re not launching missiles here):

The moral of the story here is that this unit can be used to economically supplement your tracking of freight, high value packages, etc. Just Thursday a customer I visited in El Monte asked if he could use them to track shipments — Given the fact that my trunk is probably thicker metal than the typical trailer, I can say that it should work reasonably well.

Remember these units work for up to 10-15 days in “ping only” mode, and 3 1/2 days at 2 minute updates. Inexpensive external USB-connected batteries work well to extend the life up to a month or two.

This device is very reliable, and easy to use for many security, freight tracking, and occasional tracking needs. Just don’t ask us to sell it to you to track your spouse or kids — we strictly sell for B2B (Business to Business).

Here’s one more picture of the 2 minute EZ-1000 tracking (blue with green movement/speeding dots) vs. “reality” at 10 second updates (red) — it caught me speeding… I wanted to get home in a hurry obviously:

Thanks,

Rob.

So I talked about the “inaugural run” with the EZ-1000 here.

I mentioned that we can sacrifice battery life for accuracy/frequency, and vice versa.

So on the way home from Mexico, I put an EZ-1000 in 8 second update mode and it ran for 16 hours before the battery ran out.

Once I came home (via the same route I run typically), I went for a run, & took the unit with me.

I want to show the accuracy of this unit in “high accuracy mode” — it’s amazingly accurate.

Here is the route of me driving home with the unit in my cup-holder:

You can see that I ran to the edge of the road, and purposely waited 10 seconds to begin running back to ensure an “idle” (blue) dot showed up where I stopped running:

Note that there is no sidewalk at that area of the road so I am on the Eastbound side, opposite where I drove earlier (the line without any “pins”).

Zooming down, the accuracy is ridiculous — you can see where I cross the street because the sidewalk ends and there is more room to run on the other side (both heading out and coming back) — you can even tell that I stay to the right of the sidewalk each direction!:

So it was ridiculously windy (really, it wasn’t that I just got back from 4 days of eating & drinking in Mexico and am out of shape), and I had to walk for a block twice during my Westbound (into the wind) return to home. It’s trivial to pick that out based on the map points:

And I wanted to “simulate” a patrolman taking a shortcut because they were too tired to go the full route to the entry gate & simply cut through the drainage culvert (we don’t have sewers here in the desert…):

While I was “simulating” a lazy patrolman & taking the shortcut, for the blog’s sake, I had to walk through some shrubs where I thought “I should probably not do this since it’s beginning to be rattlesnake season” (Spring & Autumn are when the snakes are warm but not too hot to head to the mountains in AZ).

As I was heading back to my house, sure enough, I saw a rattlesnake in the road (warming up in the late afternoon) but didn’t have my phone to take a picture for the blog’s sake, and to convince my kids to stay out of the desert. I ran home, grabbed my camera, ran back, but by then it was gone.

Here’s that activity:

Sorry, no fun rattlesnake picture for the blog except how long it took me to get home (.10 miles), grab my camera (and my kid who I figured I would have come with to see it), run back (.10 miles), and find the snake gone (I saw its skin it left in the dirt nearby the next day, for what it’s worth).

5 minutes, from 3:45 to 3:50 PM, according to GPS, is too long to expect a rattlesnake to sit around and wait for a neighbor to come back and take its picture (or club it depending on the neighbor…) :

Anyway, this should give you an idea for how accurate the EZ-1000 is for piecing together a security person’s day, a police-officer’s beat and/or pursuit/apprehension (pardon the pun), etc.

We look forward to helping you with your exact requirements using the EZ-1000 (or any GPS Insight offering), and hope you avoid the rattlesnakes.

Rob.

I haven’t been out golfing for a LONG time — it’s busy here at GPS Insight.

But the head of marketing for a great local company we work with (bluemedia — www.bluemedia.com) invited a couple of us out to golf the other day.

So I got to enjoy the day on Wednesday & we played 27 holes.

But I needed to test accuracy and battery life on our new EZ-1000 so I put two of them in my pocket during the day. Here’s what they look like:

One was set for 15 second updates, sent every minute (blue), and one was simply set for every minute (orange).

You can see slightly more information on the blue track, and they are slightly more accurate since they are keeping constant GPS, but again, as I’ve mentioned, that drops the battery life down. We get about 20 hours at 15 second updates once a minute. We get around 40 hours by transmitting every 1 minute. The customer will have the ability to modify the settings “on the fly” so that the battery vs. information/accuracy tradeoff is in their hands.

This will be useful for foot patrols, dismounted police and security officers, and similar applications.

You can get a lot more information from this, such as we got off on time for our 7:36 tee time, and that I went to the water to try to get the ball I rolled into the pond on the 13th hole:

This shows my vehicle parked at the golf course for 9:45 (I didn’t get a lot of work done that day…):

But this is more telling of what activity occurred there:

We are going to start encouraging our customers to consider this product in addition to vehicle tracking, particularly if their drivers need to spend extended periods of time outside of the vehicle. They will double as a panic alarm, since they have a button for this on the side.

For more information on the EZ-1000 give us a call, or look on the website and in the blog under articles in the category “hardware EZ-1000″ — for a picture check out this article here.

We are testing out this new tracking device which I mentioned in the past couple articles.

I am trying to lose a few pounds gained from the past 2 trade shows in Dallas (mmm, barbeque…).

So I went on a run today and put the tiny device in my right shorts pocket and headed on my usual route.

Here is the 2 minute-by-2 minute track:

As always with GPS Insight, green means moving, blue means idling (e.g. 0 mph as I start/end the run), and the color and thickness of the line goes from “light and thin” to “dark and thick” to help indicate direction of travel which helps visually determine the vehicle’s (or runner’s) activity more easily.

Here is a different perspective which shows me leaving my subdivision and heading EAST 2 miles then turning around at the main road. Note that the light blue line is much closer to the sidewalk than the line coming back (about 100 feet off). I took the same sidewalk the whole way — what happened? This is where battery life vs. 100% accuracy comes in. 100% accuracy, we are good for 10-12 hours on a charge. But we can go well over 3 days on a charge at 2 minute updates if the GPS unit is allowed to sleep between those two minutes — this generally adds some error to the location, in this case up to 100 feet when I’m not faced toward the satellites.

Well, as I mentioned, I put this tiny tracker in my right pocket and headed East. When I headed back, my body was “in the way” of the GPS signal from the predominant GPS satellites in this area at this time, in the Southern sky. This means they caught a bit of a bounced signal from the houses, trees, etc. to my right. This subtle difference shows up on the map. The accuracy of these units is very good (best case, 5 feet, worst case, 100 feet), considering the GPS satellites are 22,370 miles in space, and the tiny device is in my running shorts pocket (don’t worry, we won’t sell this one, and they are VERY waterproof, fyi…). The moral of the story is they’re highly accurate if you only need them for 10-12 hours on a charge, and pretty accurate if you want 3 days (at two minute updates). If you want less frequent updates, they’ll run up to 10-15 days on a single charge.

Our more powerful vehicle-mounted GPS tracking devices are always within 5-15 feet while driving, and typically within 10-30 feet of accuracy when parked. Either way, that is VERY good considering the standard accuracy of a single GPS fix is 9 meters in any direction. This is because only the US Military gets the “100% accurate” signals — for the rest of us it’s only guaranteed within 27′. However, the more samples you take in a given period of time, the more accurate you can become by eliminating the really far off samples. This is a topic for another day. Thankfully, our customers aren’t harvesting grain or painting road lines based on our “civilian” GPS. We don’t cost nearly as much either (trust me).

So to use Google Earth to its fullest, I outlined the path I ran and measured it to be 4.59 miles (I’m getting tired as we speak…):

[Nice mini-racecar track my neighbor has, huh? Unfortunately I don't know him...]

I can run a quick activity detail report for my rough speed while running (along with a “12 MPH” which is bad — it’s due to my going into the house & losing GPS signal which can cause a transient issue with this unit (currently, we are programming it to be less likely once we sell them).

Anyway, that’s some of what you can do with our new small, rechargable EZ-1000. (shown below):

By the way, did I mention it was 85 degrees today here in Scottsdale…? A little hot for a run already (sorry everyone back in the Northeast & Chicago)

Thanks,
Rob.

My last article detailed how I took a very small tracking device to the zoo to track my day there. Incidentally, that tracking device has been reporting every 2 minutes for the past 36 hours (I’ll let you know when its rechargeable battery finally needs recharging later).

I want to quickly illustrate how fast it is to overlay a map of the zoo (my son brought one home & I scanned it) in order to determine how much time I spent at a particular area (my daughter loves the Giraffes so we went there first).

After importing the overlay in Google Earth, we stretch it to fit the roads, lakes, boundaries, etc. This is covered in other “Overlay” subject blog articles.

We can measure (using the Pro version of Google Earth) that the zoo is roughly 90 acres:

Dialing down the “Opacity slider” we can see through the overlay to the “real map” to determine the distance between the real Carousel and the map carousel — only 139 feet off. This map is not military grade but it works for our purposes. We can also see the Giraffe just above the carousel. They are my daughter’s 2 favorite things at the zoo, we start our trip with the Giraffes then end with the Carousel before we leave (and a toy from the gift shop to avoid a scene…).

Here we can zoom down and bring in the tracking history of our zoo visit:

It is easy to see we were walking through the zoo at 13:31, then got to the giraffes at 13:33, left at 13:41, took a wrong turn finding the entrance to the Amphibian exhibit, then got inside the Amphibian exhibit at 13:45. Well, if you use our product regularly, trust me, it’s easy to see.

You can create geofences around key areas, set up automated alarms to notify someone via email/cell SMS message whenever someone goes inside those zones, and can set up automated reports showing how much time is spent over a week at the various areas of the zoo (or your business, or the companies/subdivisions which pay you for servicing them, securing them, visiting them, etc.).

Obviously I like to use “personal” non-business related examples to illustrate what can EASILY be done using GPS Insight. I hope you can draw some analogies to how to use our product in your business in order to save money/time, reduce risk, increase efficiency, and understand and investigate/defend employee activity.

Here is a “zoomed out” version of the Zoo map overlay — you can use subdivision maps, Army Maps, Farm maps, or really any overlay useful to your business as a guide to interpreting data, quantifying activity, and identifying points of interest relative to your vehicles’ or employees’ activity.

Thanks,
Rob.

We haven’t started officially selling the GPSI-1000 yet, but I decided to test it today on a trip to the Phoenix Zoo with my family.

The GPSI-1000 is a tiny GPS tracking device which is battery operated and can be worn on a belt by security personnel, and other generally “outdoors” workers whose safety or productivity needs to be tracked by management.

It is TINY — only 1.4 inches wide by 2.5 inches long:

I wore it on my belt under my shirt all day & tracked my whereabouts every 2 minutes.

It isn’t something we recommend for tracking vehicles — our GO-3000, GPSI-4000, and LD/HD-3500 are the best for that.

But if you want to track your employees when they are OUTSIDE of the vehicle, it’s a great way to go. It will be available later in March.

At 2 minute update rates, it’s enough to know where a person has gone, and it has enough battery to last at least a day between charges.

Additionally, there is a single button which will allow us to implement either a panic alarm or a “mark my location” capability.

Here is a map of both my tracked vehicle (Navigator, the red lines) as well as the GPSI-1000 (worn while I drove, in blue):

The GPSI-4000 in my Navigator is fantastic at tracking the vehicle with fantastic accuracy. The GPSI-1000 (person tracker, really) does a good job but is only 2 minute updates, and occasionally loses track since its internal antenna is inside of the vehicle where it doesn’t get sufficient GPS signal. That’s not its job though.

When I get out of the vehicle, then it shows where we went, every 2 minutes. I won’t bore you with details of how much time we spent looking at giraffes, zebras, amphibians, etc. But I could — all we would need to do is overlay a map of the zoo, put up some geofences to match, and run a landmark report.

Instead, I can illustrate a little of that. Bear in mind that our typical GPSI-1000 “tracked individual” might be a security guard, a dock worker, a temporary worker who doesn’t need a wired GPS device in his or her car, etc.

But I can show you how accurate and useful this hardware can be, especially when combined with a vehicle tracking device:

Basically, once the driver “dismounts” the vehicle and does a patrol, some mowing of yards, movement of equipment, etc., you can still track that individual.

And if you want to know when and for how long they rode the Merry Go Round, you can use our typical GPS Insight capabilities:

Create a polygon around the satellite image of the Carousel (giving it a few more feet to make sure we catch the activity around it):

Then cut/paste into GPS Insight:

Then Paste into “Quick Import” under the Mapping: Landmarks menu:

After quickly refreshing the menu so that the new landmark shows up, run a landmark report:

.07 seconds later:

And here is the accuracy of the units while I was on the Carousel with my 3 kids and wife:

And here were the kids on the Carousel, oldest to youngest:

And on the way there with my family, I kept it under 68 (in a 65 MPH) on the trip there & back — the accuracy between both units is very good in terms of location, speed, and time — here are 2 points right next to each other & only 1 mph off [I called this unit the "Pants Patrol" since I was going to wear it on my belt, in case you were wondering...]:

These units should be available late March for $350 (with applicable $25 quantity discounts at 10, 25, 50, 100, 250 levels). Monthlies will range from $15-40 depending on update frequency (on 1 per hour/on demand to 1 minute). We should have a heavy equipment version of this unit available in April as well.

Thanks,

Rob.

GPS Insight recently added functionality ensuring that reports are 100% accurate when running them across daylight savings time (DST) boundaries. We are headquartered in Scottsdale, and Arizona doesn’t “celebrate” daylight savings time which is actually really convenient from a computer standpoint (in other parts of the country, when DST hits, scheduled computer jobs either fail to run or run twice when scheduled between 2 & 3 AM). But since Ben Franklin invented it, & the rest of the world (and most of our customers) need reports accurately after they change DST settings twice a year, we needed make them happy.

This is something surprisingly difficult to make work for all our customers, but we finally spent a few weeks working on this to avoid customer questions (and complaints) which tend to happen twice a year during time changes.

We have always supported different time ZONES so that the same vehicles can be viewed from the perspective of user-based time zones (see a national customer’s various users with different time zones, below:)

However, when running historical reports such as stop reports for a time of year DIFFERENT than the current time zone setting, everything was off by an hour.

No longer! Thankfully, we usually only had two or three complaints about this each time it changed, but this is the right thing to do, & now everything works properly.

Bear in mind that customers running scheduled reports would always see the correct report time, providing they ran it before the 2 AM “switch” time. Unfortunately, for multi-day reports spanning that time zone, things will look weird. But that’s due to an unnatural shift from 2 AM to 3 AM and back from 3 AM to 2 AM once a year.

Nothing I can do about that, sorry — that’s where our support can help you properly interpret your reports, and we encourage everyone to call if/when you have questions, problems, or suggestions.

If you don’t like DST (like me — it puts us 3 hours earlier than NY half the year, which means I have to wake up earlier…), take it up with Ben Franklin.

Rob.

We started working with a well respected expert in the golf industry, Bill Yates — founder of Grey Town Golf, LLC. They are a golf management company in Pebble Beach, CA. Bill’s expertise is consulting to courses about how to improve the pace of play.

Based on his interest in a prior blog article we have begun work on a system for using GPS technology to improve the pace of play as well as maintenance and service at golf courses by tracking the various types of golf carts.

Starfire in Scottsdale was nice enough to work with us on this project, and here are 5 of their carts for yesterday:

Here is that activity as a movie:

GPS Insight movie of golf carts throughout the day

I need to remember how to embed that video into the blog, but for now it will open up in a new window.

This sample data allows us to now write alerts, reports, and maps which will allow Bill and Grey Town Golf to improve their pace of play system significantly.

Here is a “current status” dashboard view of the course and the 5 carts we’re tracking currently:

We look forward to bringing this system to market in early 2009 (and maybe I’ll get some time to test it out on the course myself…).

Rob.

This is the 3rd (and last) in a series on GPS tracking of skiers, and more to the point, the overlaying of ski maps into Google Earth in order to put that data into reference. So after asking someone way better at Photoshop than me to “stretch” that ski trail map of Heavenly Lake Tahoe and make it fit Google Earth, it has become VERY clear that those maps aren’t even close to accurate.

Photoshopping the map & stretching/skewing it to fit, then overlaying the sides to be accurate, it’s still obvious that there is no real scale built into these maps. The Gondola is nowhere near reality:

So we aborted the mission of overlaying this map in favor of simply recreating the runs as Google Earth “paths.”

Here is what it looks like once you add a number of “paths” and “placemarks” in Google Earth, using the map as a reference, and the satellite photo with the missing trees as evidence of where the ski runs actually are:

The paths are color-coded based on type (typical green/blue/black difficulty based on the map) and the Gondola is in red.

If you were using GPS Insight to say, track your Snowcats (which are used to groom the runs regularly), you would be able to run reports on which runs were groomed on which days, for how many hours, etc. This is actually something we’re starting to see some interest in from some ski resorts which is part of the motivation for this exercise (if I had thought about it in advance I could have written off the trip!).

You can see how accurate this is if you take a little time to properly map the trails — We show activity skiing right along Orion, Skyline Trail, and Ridge Run (as well as us taking the lift up ABOVE Ellies — I didn’t take my 8 year old snowboarder on this black (he probably would have done better than me, actually).

Using a combination of a ski trail map and Google Earth with recent imagery, it is easy to see which runs particular GPS data recorded activtity on.

We’re going skiing tomorrow in Pinetop AZ at “Sunrise” park (www.sunriseskipark.com) — here’s their much more straightforward trail map.

I’m done overlaying them, I think everyone gets the point now (on to more vehicle based GPS tracking topics next!)

Thanks,

Rob.

It’s been a month since GPS Insight moved our corporate headquarters, and we’re pretty well settled now.

Here are some pictures of the new area and some of our employees:

Here is Ray in our tech support bullpen:

Here is Alissa in one of the primo salesperson offices:

Here is our main conference room with a projected image of all 11,000 vehicles currently tracking under GPS Insight:

And here’s Elliot trying to decide between all the high tech coffee choices in our break room:

And no one has knocked me out of my office just yet per my earlier offer:

Evelyn standing in my office (I still haven’t quite unpacked…):

And from my office, I see this coyote pretty frequently:

And since it’s getting dark earlier this time of year, this is a familiar site as well:

Any time you are in Scottsdale, please let us know and we are happy to meet with you in person at our office and introduce you to the gang.

Here are a couple other pictures from last week (Rob just got a new camera…):

Elliot at his desk:

Trent, who runs technical support:

And last but not least, (and not all, but this blog article is getting long…) here are David and Alan:

Rob.

After 3 years in our existing space, it’s time to expand (again) and this time move to some newer space. It’s just 2 1/2 miles away from our existing space, and although we’re moving to 6500 square feet from 5000 feet, we’re probably going to have to give up the golf simulator and the foosball table, although my kids (and maybe a couple employees) will never set foot in the office if we don’t bring the Wii with the 100″ projected screen. I’m not sure where it will go but I’m sure we’ll figure it out.

Here are a few pictures:

Reception:

Outside view:

Conference Room (the big reason we needed to move — ours was half as big and staff meetings were getting a little too close for comfort):

Rob’s new view (I took the worst one of course):

Where all the important meetings take place:

Where your inbound tech support calls will go — 25×18′ of (eventually) our GPS Insight tech support gurus. We have a lot of desks to move…

And the salesperson of the month gets this office:

And if they sell 2,000 devices that month, they can have my office that month… (I’ll probably be traveling anyway):

The building we’re moving into has lots of room for expansion, and so do we — we look forward to servicing our customers from this new space, and hope to see you visit us in it soon (sorry, no ‘General Store/Antiques hanging from the rafters’ restaurant to take you to lunch at in this complex unfortunately — just a coffee shop next door & we’ll need to get in our car and actually drive somewhere now).

And we can always track it — here’s the old office (me getting caught up this weekend):

And the new office (me trying to figure out where everyone will go just before I head home after driving the weekend home-office-office-home circuit):

Rob.

That sounds like a lot more important title than it really is. My family and I went to Mexico for Spring Break last week and came back today. It takes a lot less time to get INTO Mexico than it does to get back. Everyone asked us “How long did you spend in line at the border?”

So I thought I would take a quick look.

We all loaded up in the family truckster (Navigator) and left town Wednesday (after 2 botched attempts trying to leave Scottsdale forgetting, in order, the radar detector, and then our passports [note passports aren't necessary but I didn't want to find out that changed recently]). We came home today (Easter Sunday) and had to leave Rocky Point early since we’ve heard about the wait on Sundays back to Arizona taking over 4 hours.

Here’s a quick 10 second map-based history which I ran:

Here is where we parked the car for a day and a half before venturing out to town to buy the obligatory tourista materials each year (maraca’s, mini-guitars, vanilla, tons of shrimp, and the obligatory “Cuban” cigars…).

Zooming down on where we stayed shows where I parked the car for a few minutes while checking in (beautiful property, by the way, called Las Palomas at Rocky Point, Mexico) — note that while we drove 257.5 miles up to that point on Wednesday, the last “leg” of the trip (after getting lunch) ws 103.1 miles. By the way, I would have lost my sanity on this as well as all other long trips if not for Sirius Satellite Radio and 2 DVD players for our 3 kids on the way down…

Now for the fun part, heading back to the border — unless you want to get stuck in line for 4 hours, you need to leave early in the AM. With one screenshot, I can tell the following:

* We left at 8:12 AM (first point under the March 23, 2008 folder)

* We stopped for breakfast from 8:22 until 8:35 getting some doughnuts and coffee for the drive [note the blue idle dot, yellow “off” stop, and second blue “idle” stop from 8:22 until 8:33. There was one person in front of us in line — all I can say is that things move slowly south of the border…

* We went under the posted speed limit numerically (although our units of measure were off) — all of the light green dots show “Speedy Gonzalez activity” over 67 mph. The one I clicked upon (86 MPH) was technically below the limit of 90, but that was KPH…

* Not all of our maps support non US roads (not really a problem, since we won’t actively sell to non-US customers until Q3 2008)

* We spent less than 40 minutes from the time we arrived “in line” at the border until we got past it and back on our way — the first “clumped together” point has a 9:44 timestamp on it, and the first point across the border was at 10:24.

Let’s quantify just how long it took getting INTO Mexico vs. coming back from Mexico by drawing a quick polygon geofence around the border area:

Running a quick report with a couple mouse clicks looks like this:

Yields this report, which shows us that it took 2 minutes to clear the border going INTO Mexico but 40 minutes coming back. This report only took .12 seconds to run and also shows us that we spent 3 days, 18 hours, 9 minutes between border visits:

Because we use AT&T/T-Mobile’s GPRS network, we lose contact with the vehicle once it’s over the border, and will show you how quickly we forward that data back once we regain coverage. The “vertical” lines show long “lags” between a point being recorded & it being reported via cellular transmission.

For the entire trip, the vast majority of the points are into our database and available to the end user/customer within 3-5 seconds. Predictably, where we stop network compatibility, we stop reporting in real time — however, the second we hit coverage again at the border (typically, a couple miles prior to the border) all of that data is forwarded up within seconds so no history is ever lost:

In reality, we forwarded up all of that data at exactly 9:48:29, 1/3 of one mile South of the border:

But here’s a better idea of what is going on .31 miles south of the border while waiting in line to get through — everything imaginable is being sold, none of which would fly with my homeowner’s association. My favorite was the gold painted grim reaper:

I was told it was almost free. Only 5,000 pesos. Maybe next time I’ll pick it up if I stay too long and wind up spending 4 hours in the line at the border. Today we were OK with some fresh tortillas for 2 bucks.

Rob.

About once a month we hear from a customer who has recovered a stolen vehicle using GPS Insight (the other day we heard from a Miracle Method franchisee which led police to a recently stolen vehicle and had it recovered in >> SIX MINUTES <<).

But honestly, that’s not what our customers purchase GPS Insight for — it’s a nice side benefit, but most companies purchase in order to gain efficiencies, save money on fuel by curbing idling, off-hours usage, etc.

Here’s an out of the ordinary story about finding a lost vehicle using GPS Insight which occurred last month:

I paid a visit to a Las Vegas tour company last month because I had 4 days in Vegas for a conference & wanted to visit some of our customers there.

One of this company’s tours takes passengers to an extremely remote rim of the Grand Canyon.

The same day I visited, one of their vehicles got stuck in the mud well outside of cell phone coverage, and used their satellite phone in order to ask for help. They stated they were 35 miles “onto the dirt road” but really didn’t know where they were.

Unfortunately while our GPS tracking works just about everywhere since we use AT&T and partner (T-Mobile) coverage, we had lost coverage of the vehicle right where the cell coverage ran out, so using GPS Insight wouldn’t help in this situation.

Or would it…? I asked the manager when the last time another vehicle took that route was, and he told me back on the 5th of February, a particular vehicle ran the same tour on that same remote, dirt road.

I ran the history for that day for that truck using our interface like this:

The scenic history of that vehicle’s trip to the North Rim of the Grand Canyon is shown below (along with the fact that the tour remained there for 1 hour, 36 minutes):

Then I used the path distance measuring tool in our 3-D mapping platform (Google Earth) to measure exactly 35 miles from where the road turns to dirt. That placed the “in trouble” vehicle exactly at Craig Ranch Road:

They were then able to send the towtruck to that remote intersection and know that they would find the vehicle. It was getting dark and the group of tourists had been stuck in the mud instead of viewing the Grand Canyon. This was the first “breakdown” this well-known tour company has ever had. Thankfully I was able to help them significantly narrow down the search area.

As much as I hoped to hear things would go well, I had a meeting to get to, and left, then found out the next day that they were able to find the truck and everything wound up good.

I was curious how close I was to accurately defining the vehicle’s location (or really, how accurate the driver was when he said they were 35 miles onto the dirt road), so I pulled all history for the day of my visit, and see that we were only 2.3 miles from where the vehicle was actually stuck.

I can also see at a glance what time they got moving again (the green (moving) points in and out of the “stuck” area have the time right before and after the long stuck-in-the-mud experience) — from 12:44 to 18:03 — roughly 6 1/4 hours.

I like to think I helped that tow truck to find the stuck vehicle, but even if we hadn’t, GPS Insight provides a great way to document exactly where and when that vehicle got stuck, and what time it was back on its way home.

Additionally, we are able to show where the AT&T coverage stops and you can see that the coverage extends almost all of the way to the end of their trip. It is very rare to lose AT&T coverage for our customers, but if and when they do, we can very precisely determine where the remote fleet vehicle does or does not have coverage. The height of each “pin” shows how long the vehicle was out of range of cellular coverage. What we call “lag” is how long it takes for the data to get to our database from the time the unit transmits — note typically it’s only 2-5 seconds, sometimes 10, when in good cell coverage, which is 98% of the time for all of our tracked vehicles.

The truck was out of cell coverage for 25 miles as shown by the measurement below.

Bear in mind that where we are in coverage, it’s typically only a few seconds from the time the vehicle reports its position until it’s available to our customer. This vehicle goes to some pretty desolate areas.

Additionally, bear in mind that using satellite communication to transmit vehicle location costs >> SEVENTY TIMES MORE << than cellular costs. A typical 1 Megabyte plan is $10 per month. A typical satellite 100 Kilobyte (1/10th of 1 megabyte) is $70, last time I checked. That’s why this customer doesn’t mind losing track of its vehicles for this short period of time/distance. They know where they’re at since they know when they went out of range, and on what road. Once they get back into coverage, all that data is sent back to the servers, so things like speeding, stop duration, excessive idling, and engine trouble codes are all “stored and forwarded” for analysis/alerts/reports later by management.

Some day I may have time to take one of these tours, I live in AZ and have for almost 5 years, and still haven’t seen the Grand Canyon other than from an airplane (which is about every month as often as I fly over it).

This was a long article, thanks for reading it.

Rob.

In an earlier article I mentioned I would do a blog article about the GPS Insight route report. Here I will create a simple route (from my house to work) and run the report to see which of our vehicles ran that route. It should only be (and is) both my vehicle as well as my wife’s, since she sometimes comes by the office, or I sometimes take her car to work (I did the other day).

We construct the route by creating, then selecting a begin & end landmark, in this case “Rob House” and “GPS2″ :

Then we run a report for a range of dates, a group of vehicles, and a particular route:

Which yields this report, which shows that the only two vehicles making this trip between Rob House and GPS2 (our office) are mine (Rob) and my wife’s (Navigator). I have put red boxes around strange entries, where I either leave my car at the office for days (e.g. someone drops me off at the airport), or it takes me a long time to drive the 6.6 miles I live from work (e.g. I go to a customer meeting before heading to work), or it takes me a while to get home from work (e.g. I go out to dinner with a customer or my family).

At the very bottom I put a red box around the only time I’ve used the Navigator to go to work, this being because my wife took my car to the airport for a trip and left hers so I can take the kids to school/soccer/etc.

Perhaps I want a “median” time it takes me to get to work, and how long I usually stay at work. In this case, I would pull the data easily into Excel like this and run whichever custom calculations I need to:

Now I know it takes me 19 minutes median for me to get to work (the middle of all the samples, sometimes better than an average), and 10 minutes median to get home.

This is because I typically drop my sons off at school on my way to work, which takes around 10 minutes, but I come straight home.

This is a simple example of how you can use GPS Insight to calculate route statistics and use them to gain valuable insight into your fleet and your drivers.

Thanks for your interest, & call us at 866-GPS-4321 to talk about how we can help your company to do similar analysis with its GPS tracking project.

Rob.

I was talking to Jerry, my pool guy, the other day while he was servicing my pool in the morning, and he has been using GPS Insight for some time. He has a customer who complained that he hadn’t been to her house while she was out of town, and she said she was canceling and not going to pay. Jerry owns AZ’s Best Pool Service and does a great job, & does half of my neighborhood based on my referrals, so I know he wouldn’t skip a customer.

He was able to prove service in the following way. He remembered he goes there Thursday, and she was gone sometime in the middle of September, but not which truck he would have sent or used. So with a couple mouse clicks, he ran a history report for Thursday the 13th of Sep through Thursday the 20th:

This yields the following map (after selecting only the first/last days of service and turning off roads to see the two vehicles’ paths more clearly in Green/Blue):

We can easily adapt the map to show ONLY stops > a certain number of minutes (e.g. 5 minutes) and then zoom down to the neighborhood and see several “pairs” of stop pins (in yellow, which indicates a stop less than 60 minutes):

Clicking on one of the two stops shows the number of minutes stopped (13) and the exact time to the second that the vehicle was turned off. For space reasons I only show one of the two stops. With this information, Jerry can easily prove that he attended to his customer’s pool.

Another way to prove a service is to use a circular or polygon shaped landmark.

I can show the amount of time Jerry spends at my house on my pool for the month, to include last Friday, which should be a longer than usual stop because we were talking about this & pulling it up on my office computer since he wanted me to see how it helped him. Or better yet, we can draw a polygon around our neighborhood like so, and run a polygon inclusion report which will show the visits and the amount of time between visits:

We run a polygon Geofence report like this:

And can see just exactly when Jerry (or one of his employees) drive any of their vehicles into and out of our neighborhood, how long the visit was for, and additionally how long “between” visits:

I highlighted the trip from the other day where I talked with Jerry for 20 minutes, which slowed him down and made his neighborhood visit longer than the typical. He obviously had some single visits scattered throughout the month, and if you wanted to do things like average/max/min stop times, you can easily export the data into Excel by clicking on “Download Now.”

This is one of the best aspects of the GPS Insight product, and GPS tracking in general — service companies are always able to prove (or disprove) service in order to ensure the customer is being taken care of, and that additionally, billing which is due can be proven. Comparing historical times of service can be used to determine how efficient a driver is, or alternatively, if that driver is not taking the necessary time at each of his or her stops.

Rob.

We have recently begun to work with the ADOT (Arizona Department of Transportation), and were asked to incorporate mile marker data into GPS Insight for them. Of course they have this information somewhere, but in the time that it will take to get the raw data and process it into GPS Insight, we could simply create these mile markers “manually” and I thought this exercise is worth documenting on the blog.

First, I found a high resolution map of the mile markers online via Google.

Then I took a close-up screenshot of the map for the I-10 [I just traveled this last week myself so I thought we could do some interesting analysis of this data later].

Now I can import this “overlay” into Google Earth and use it as a template for creating the mile markers:

Note that the important thing here is that the part you really want to “trace” is accurately stretched over the map — maps and Google Earth inevitably diverge, since maps are never as accurate (or spherical) as Earth. The other roads/borders will not match exactly unless you change the “keystone” using a graphics editor. It would be nice if Google would add keystone as an editable aspect of overlays in the future (keystone is effectively stretching an image more on one side than on the other).

Now that we have this image, we can create landmarks on top of each “10″ mile marker (the map only shows the “tens”).

Since this particular mile marker is on the edge of what I considered “accurate” (due to keystone) I measured it myself using the Google Earth Path ruler to 10 miles from the border.

Then a quick copy/paste/rename/replace of the original allows me to quickly place the next 14 “tens” mile markers, placing them on the red dots but on the road itself from Google, which is always far more accurate:

In order to move a copied/pasted landmark, make sure the original is “turned off” (otherwise you can’t “grab” the one you want to rename/remove) and then right-click it, choose properties, and then you can move it when your cursor turns to a pointing finger. Remember to change the name (from 10 to 20, 20 to 30, etc.).

Here is the final product (note I created a folder in “my places” and placed the numerically sequential points along with the overlay map):

Then for an additional I-10 definition I will draw a rough polygon landmark around the I-10 which will help me to answer questions like “how much time do my vehicles spend on the I-10 between mile marker 0 (border) and 150 (Phoenix)?”

The key to accurately defining borders around large areas is to use the “compass” area to navigate between clicking between areas. Plus you can right-click to remove your last “incorrect” points and zoom in/out, move, and pan/tilt using the mouse at the top right, since your mouse otherwise is being used to define the polygon.

Now we can easily find a particular mile marker by typing “ctrl-f” for find, then “I-10 40″ to instantly find/double-click that point for instant navigation:

And last (most importantly), you can instantly answer questions about your fleet like “How long did Rob (in the Navigator) take driving to, then back from DisneyLand last week on the I-10 in Arizona between mile markers 0 and 150?”

We see it took me 1:50 to get TO Disneyland on the I-10 and 1:41 to get back during that same stretch of road, and that it took roughly 3 days and 7 hours from the time I left Arizona to the time I returned (since it is on the border at Mile “zero”), by running this report (which takes a split second to complete):

And here is the actul trip itself along with a 6 minute stop (we were traveling with our kids…):

You can see we made it a whopping 83.2 miles before hitting the first of what was many rest stops on our way to DisneyLand…

And that stop was apparently at mile marker 87, 3 miles before 90, between the 80 & 90 we just created:

Thanks for reading through this practical example of how to use GPS Insight and external mapping resources (ADOT mile marker maps, Google Earth, etc.), along with real GPS data to answer questions you may have about your fleet. Of course ADOT would want to answer other questions (response times to accidents in certain areas of the state, proportion of their vehicles in any one particular concentration) but having these landmarks and polygons in the GPS Insight system helps them (and all of our customers) to answer these types of questions.

Here’s the Google Earth file for you to see the results (minus my DisneyLand trip):

http://www.gpsinsight.com/tmp/ADOT Mile Markers.kmz

Rob.

In this article I will detail how I use GPS Data to answer questions about cell phone signal inadequacy in Arizona using all of our data collected in August.

We have a large municipal customer who commands some respect from AT&T and is able to actually ask them to put up new cell phone towers in certain areas which are lacking.

They track a large number of vehicles using our GPS Tracking product, and asked us if we could help them determine the worst cell coverage areas in Arizona, based on our customers’ data.

Since we track several thousand vehicles in Arizona, we were able to do this.

Above is shown a real-time location of all several thousand vehicles we track in the Arizona and San Diego area.

I took the data from August so far and compared the track time with the time we received the data, where the “lag” is due to cell signal being unavailable.

We pulled 2500 points ranked by longest to shortest lag, where the shortest lag was 6 minutes — bear in mind our updates occur in 2 minute intervals and typically take 5 seconds to make it from the tracking device and another 10 seconds to be processed and pushed down to your map.

I plotted these lagged position updates and color and height coded them based on how long they took to ultimately get forwarded.

This shows all places where there was some level of poor cell coverage in Arizona for August:

Thankfully, when comparing where the vehicles we track are, relative to where the “dead spots” are, they are not very troublesome, and do not cover where most of our drivers tend to be:

Then we can place it in a 3-D view where the height of the lagged positions indicates how bad that coverage is (e.g. how long the vehicle was out of range after it collected that point):

Then I can pull both a Cingular/AT&T Coverage image I prepared (from their website) and overlay it with Arizona. This map is not 100% accurate, and there are a lot of T-Mobile and partner network areas where our product works which are NOT depicted, but it gives a good idea of where our coverage is (it follows the roads and major metro areas typically. Note the edges are not exactly aligned since our mapping knows about curvature of the earth but the Cingular map does not:

Zooming down we see that the worst area of coverage is not “orange” (which indicates areas AT&T claims to work in):

Zooming down on this area, we see it’s a desolate road leading to a mine:

Looking at the data points relative to the aerial photo, we see how accurate our GPS Track is. Then we can zoom WAY down and see the exact areas of the mine which are being driven in.

So if I was Cingular, I wouldn’t worry about covering this remote mine, but I would worry about coverage where I pointed out with 2 arrows below, and perhaps place or adjust the cell tower there.

Knowing there are around 800 square miles in all of Arizona where our typical vehicles (e.g. forgetting about the mining vehicles) go out of range for more than 5 minutes is great to know, since there are 111,500 square miles in Arizona.

This means we are out of range less than 1% of the time. Since most GPS tracking products (ours included) store this data until the vehicles return to coverage, there is never any data lost. We store roughly 45 days of typical driving, so if a vehicle goes down into Mexico (where we do not currently have service) and back, we still will be able to show the exact locations they were at, if they were speeding, etc, once they return.

One of our customers drives back & forth to Mexico and is currently using our service to include our polygon geofences in order to ensure compliance with Federal law, and to check their times spent at the border (which helps them in their business due to certain reporting requirements). I will cover this company and their unique usage of the GPS Insight product in the future.

This is the first post to go into this new GPS Tracking, Reporting, and Mapping Blog. The intention of this blog is for me to share my ideas, experience, and experiences with GPS tracking with the fleet community as well as the GPS Insight customers and employees who frequent this blog. I will make it a point to be product neutral and strictly “positive” about other products in the marketplace, but of course any examples or screen shots will likely be from our GPS Insight product, with more information available at www.gpsinsight.com. Comments and questions are welcome, and the overlap on this blog with both wiki.gpsinsight.com and forums.gpsinsight.com will eventually be apparent as content gets duplicated to all of the appropriate areas. Thanks for your interest, I hope this blog will help you, as well as help us to help you…