search

BMMRO-tech / BMMRO

MIT License
3 stars 5 forks source link

Spike: investigate iPad GPS technology #290

Open njktw opened 2 years ago

njktw commented 2 years ago

We want to understand the iPad's ability to record GPS coordinates so that BMMRO can accurately reflect the encounter.

When the BMMRO team are out on the boat and they encounter an animal they have to log the location of the whale for the duration of the encounter. They do this to build a pattern of how the whales move around. We need to understand if the iPad is able to robustly and consistently provide the level of detail they need to do this.

In scope / Key questions we want to answer:

  1. can Apple's GPS record location to the required precision (within a few metres)?

  2. how many decimal places is 'a few metres' in terms of GPS position?

  3. If 1 is not possible end the spike and add the findings to this ticket. Suggested course of action is to remove automatic population of GPS from the app and use the manual marker instead

  4. If 1 is possible then [next spike tbc] is in play

zeingowie commented 2 years ago

Answer to number 2 would be 'between 4 and 5 decimals'

Screenshot 2022-06-23 at 00 14 16
njktw commented 2 years ago

Meeting with Charlotte BMMRO - Unnamed Area (3)

Meeting with Charlotte BMMRO - Unnamed Area (4)

njktw commented 2 years ago

As per the screen grabs above - we followed up on the GPS issues and broke it into 6 threads.

The first 4 are about the refresh reliability.

The last 2 are about the accuracy of the coordinates, which can also be expressed as trying to explain the variation between the reading Charlotte gets in the App and the reading she sees using the Garmin handheld GPS device.

The second image illustrates the impact of a moving boat. We established that if the BMMRO boat is travelling then the coordinates will vary by differing degrees based on the speeds involved. This is because of the 10 second window the App has to establish a connection to the GPS network. If the connection takes 8-10 seconds and the boat is moving then there will be a material difference in the coordinates between the App and the Garmin.

The result of this investigation was to explore other options with BMMRO to do with the available data from the Garmin handheld GPS device. The device outputs a csv dataset of the waypoints in it's log book so we will explore avenues to integrate that data into the import scripts and queue up a story to address this