The Intel Open Source Technology Center UX UI team got the mission of contributing to improve the user experience for controlling interfaces belonging to the Dronecode Project, making them accessible to both new and advanced users.

This mission started with the team doing extensive research on Ground Control Stations (GCSs). The results of this research showed that a redesign of the compass calibration process was needed. Thus, this project started.

The ArduPilot flight stack (which was a part of Dronecode at the time this project was developed) has an embedded compass calibrator. During the calibration process, samples collected are distributed over a sphere to get calibration parameters.

In May 2016, they introduced to their compass calibrator a mask representing sections of a geodesic grid - this mask provides a picture on how the calibration is progressing ang guide the user when they are performing the calibration.

The fact that the embedded compass calibrator provides a consistent behaviour on different GCSs and that a graphical solution was already available - which is important in Open Source - made it a priority to incorporate this solution in our redesign for compass calibration user interfaces.

The Intel UX team conducted a usability test on QGroundControl, which is part of Dronecode. The main purpose of the test was to observe final users dealing with the QGroundControl interface in order identify gaps and opportunities from a realistic point of view.

The QGroundControl usability test revealed that users didn’t know how to start the calibration process, didn’t know how to move to the next steps of calibration, were not sure if the process sequence was a must and found the feedback unclear. With this in mind, the compass calibration redesign was started.

The first step of this redesign was to define the task flow for the calibration process. The task flow for QGroundControl compass calibration was used as a reference, as the problems identified in the usability test originated from the confusing task flow and feedback.

The new task flow aims to make the user trust and feel secure in the process, as it guides them from set up to the result of the calibration - there is no doubt if the process was successful or a failure.

The calibration phase - when the user has to rotate their UAV - aims to teach the user how to rotate it with an instruction animation and the live geodesic grid being completed. The user gets instant feedback to feel secure on the process.

The results show success or failure - based on the GCS parameters - as well as specific numbers of interest to advanced users.

The interface is based on a grid divided in 3 columns, even though the screen is divided in two parts. The purpose here was to make it adaptable to future tweaks and to guide open source developers in the case they needed to change it.

The instructions for calibration are given by an animated UAV 3D model, while the geodesic grid is completed accordingly to the user’s action with the drone. We divided the screen in 2, so the user can discover and manipulate the geodesic grid.

This design is meant as a guide for developers, so we left escape options and the introduction of a possible menu open. The solutions are proposed and the assets are made available so developers can implement their solutions

We decided to use an animation as it presents clearer instructions - following the goal of reassuring the user from start to finish. The animation presents one movement at a time, so the user doesn’t need to wonder about the sequence - and can learn to rely on the geodesic grid.