The link below shows an example of 5 wineries in the Yarra Glen region using Google maps. The same data has been created as a scenic route able to be loaded to a TomTom device using the Add to TomTom button. Here is the link:

Yarra Glen Wineries

As part of this test I used a POI database containing petrol station data and flagged those stations that were BP affiliated, Melbourne based, 24 hours with Car washes and ATMs. Rather than use the TomTom make0v2.exe tool, I manually created an ov2 file via the TomTom wizard. The wizard outputs the necessary html code to place on the website to create both the addtotomtom button and reference the POI information created via the wizard (and hosted by TomTom).

Separately I created a JSON file with the same locations for geocoding on the fly and locating on a Google map for reference purposes. Here is the page link:

Add To TomTom POI Example

Haven’t posted here for a while. As an exercise for work I’ll use the next few posts to create some custom POI sets and Scenic routes to load into a TomTom unit. I’ll also set up a web page using the add to TomTom button which links to TomTom HOME. From there I’ll test the TomTom unit with these POIs and scenic routes in the field. The plan is to load a set of petrol station POIs with special fuel types and a scenic route for either Yarra Glen Wineries or Melbourne which includes Melbourne’s famous lanes and alley ways as way points.

Back to the Google Earth time animations I put together in [Post 028]. In this one I’ve kept the time animation sequence from the race and added an image (ground) overlay over Hawaii. The hawaii image was traced out using an image program – then I used Irfanview to save in .GIF format and set some of the perimeter colours as transparent so the image showed up in Google Earth better.

Google Earth has a fantastic set of tools to precisely position and stretch/ compress images to get the right fit once the overlay is run. Then its just a case of getting the lat long bounding box co-ordinates to reference along with the image in the ground overlay KML tags.

Here it is:

 Hawaii Bike Route on Hawaii overlay?

Here are two examples of map presentations using some code borrowed from the Google Maps team. The first example shows a full screen map presentation (which fully caters for map recentering and resizing) and the second example shows a side bar map presentation. I’ve used the Hawaii Ironman bike course as the display material:

Full Screen Map

 Map with Side Panel

Following on from the previous [post 028A] I’ve taken the excel data with positional co-ordinates and associated time stamps (mine), and dumped it into a KML file. The KML file includes both the Hawaii Ironman bike route as well as the individual positional placemarks spaced approximately 5 minutes apart in the timestamps. Here is the link to the KML file:

 Hawaii Ironman Bike Course Animation?

The Google Earth Time Animation feature has been around for a little while so I thought I’d try it out – using the Hawaii Ironman 2007 course. The results are quite spectacular – and are really a nice way to build route animations if you’re not into Macromedia Flash animations.

Basically any KML feature can have time data associated by using timestamps for individual point placemarks. I’ll be using the Z (UTC) time format structure for this example. The time slider facility in Google Earth then allows you to set playback speed and some other variables. Other polygon shapes and ground overlays can be added into the time sequence to improve the display. Timespan elements can be used specifically for ground overlay animation.

Also, there is a nifty checkHideChildren function to ensure that individual placemarks do not appear in list view.

I thought I would start with the Hawaii Ironman Bike Course and I’ll be building this example over the next few posts based on the data spreadsheet below. This spreadsheet lists out individual point placemark co-ordinates along the bike route – as well as the distance between co-ordinate sets based on the distance algorithm used in [Post 027].

Next I’ve used my own bike pace data to calculate an approximate timestamp for each co-ordinate set. In the next steps I might add some data for the race leaders (Chris McCormack and Chrissy Wellington) and add in the swim and run routes as well.

Here is the initial spreadsheet that I’ll be using to transfer data into the Time Animation KMLs: 

http://spatialities.googlegroups.com/web/Distance_algorithm_HIM_01.xls

Telstra keeps a list of Wi Fi hotspots on its website which mainly includes publicly available access points through the likes of Starbucks and McDonalds. I was interested in how many of these hotspots were located along the Nepean Highway route between Sorrento and Melbourne. First of all I geocoded the 8 applicable addresses here:

 Wi Fi Hot Spots

Next I used a nifty set of code to also show the range of the Wi Fi hotspots:  

I tested out my new Nike+ipod (Nano) kit to see how accurate the distance measurement feature was. Walking the same route as [Post 025] the Nike+ipod showed a distance of 4.11km travelled (versus the Garmin GPS measured 4.03km). Not bad for a basic RF sensor/ receiver unit, especially since I left it uncalibrated. Another toy that I can use as a training aid.  

It’s been a while since I put my Garmin forerunner 205 unit through its paces, so I took it on a brisk walk for 40 minutes around my suburb. At the same time I set up a polling schedule using a carrier network tracking service (non-GPS based) so that I could compare the accuracy of each service and the results.

The Garmin unit tracked a total of 423 points over the 40 minutes whilst the carrier network service was set to poll my mobile phone in 5 minute increments based on GMT. Back home I downloaded both sets of data, matched the GMT times up (using 5 minute increments) and dumped the latitude and longitude data into a spreadsheet.

In this spreadsheet I loaded a simple set of trigonometric functions to calculate the distance between two sets of points based on their respective latitude and longitude co-ordinates specified in degrees. ie this gave me the distance in metres between the GPS specified co-ordinate and the non-GPS service specified co-ordinate for each nominated GMT time increment.

Here is the spreadsheet link: 

 Spreadsheet

It came as no surprise that the GPS measured co-ordinates anchored almost directly onto my walking route when the results were dumped into a map. However, the distances between the GPS and non-GPS measured co-ordinates (and the distance the co-ordinate was “off” my walking route) were interesting. Four results were close (<400m) and two results were way out (>1900m). Many explanations are possible including the the usual signal attenuation “suspects” but there is also a possibility that my mobile was passed to a mobile base station which was not necessarily the closest in range.

Here is the map link showing the results: 

 Location Data Display