PRE-UX
Here I would like to show my practical experiences developing graphic interfaces as an engineer and before knowing the specific methods for this work.
They do not follow the necessary procedures for the practice, but they can show the foundation of my informal experience in the area.
The three main projects that I worked and would be improved with nowadays knowledge are:
- Helix Project
- Hydra System
- Tīma Interface
Helix Project
My first experience as a user interface developer was on the Helix project.
Helix was an unmanned helicopter-shaped aircraft; nowadays it would be equivalent to a drone (at the time, in the mid-1990s, the term drone was only used for military applications).
The aerial robot was developed by a start-up in southern Brazil for observation, inspection, and patrolling tasks.
The aircraft had an embedded Transputer (32-bit RISC CPU) that used an in-house developed operating system programmed in Forth and Assembler for mathematical processing, in addition to two 8-bit microcontrollers for input and output processing of sensor and actuator signals.
From position, attitude, and acceleration information provided by an electronic compass, a digital gyro-inclinometer, a GPS receiver, and an inertial reference unit, the controller calculated accelerations, linear velocities, and rotation rates relative to a ground reference, as well as information on the attitude and absolute position of the aircraft.
I helped develop the graphical interface for the Ground Station, responsible for mission control and flight recording. My tasks also included writing promotional material for the general public and potential investors, as well as documentation for sponsoring research agencies.
The interface was implemented in C/X-Windows on an IBM Power Station running AIX installed in a support van that also transported the aircraft to the mission location.
Decisions about the interface were made by the engineering team based on the available literature on civilian and military manned aircraft.
Thus, the mission control screen was divided into two main areas: piloting and telemetry.
At the top, for piloting, we had three subdivisions: the piloting display with information on speed, distance, azimuth, and attitude of the aircraft in relation to the next waypoint, which was superimposed on the video camera image to prevent the pilot from having to divert their eyes to obtain this information. Simple graphic figures were used to avoid noise in the image but at the same time provide the data.
Below the piloting screen, we had the Vertical Situational Display, which showed the profile of the region overflown relative to the aircraft's line of sight for the next 4 km (represented by a dotted line).
On the right side, the Horizontal Situational Display represented the map of the region flown over (at the time, I had to manually digitalize the information from a printed map).
The screen also allowed for scaling changes using the zoom in and zoom out buttons.
In the lower third of the screen, we had Telemetry Information Display, which shows propulsion information, communication information (radio link status), aerodynamic surfaces, energy systems, and satellite navigation (GPS) information.
All drawings and animations were built in C; part of the scales was linear, and part was exponential to increase fidelity to real instruments; we also took care of information architecture and grouping by themes; last but not least, we used colours to differentiate normal usage situations from those that required more attention or even urgent decision-making.
About Helix UX/UI design
At the time, none of us had studied anything about UX/UI Design, and we searched for references in the available literature in scientific articles and specialised aeronautical magazines on the practices and organisation of pilotage-related information.
It was a tremendous learning experience, especially considering all the challenges faced during the project; the team responsible for visual interface was composed for only two people: one senior engineer and me. I imagine that today we would invest much more in usability research and would have many more references to drone piloting apps to build a more user-friendly interface.
Four prototypes were successfully tested (approximately 400 flight hours), but from a commercial point of view, the company did not achieve the corresponding success. There were many technological, governmental, geographical, and even legal limitations that culminated in the use of the acquired knowledge for the development of another company; this time, automated effluent treatment with remote control and monitoring.
HYDRA SYSTEM
The Hydra System was conceived by the company Acqualan with the aim of using automation and telecommunications technologies to optimize the operational services of water supply systems and effluent treatment.
The idea was to distribute compact modular treatment stations throughout the city (dozens or even hundreds) that could be operated and monitored remotely.
This project was much simpler than Helix since the remote stations were much less complex and had a lower operating dynamic. The company built its prototype on a beach, and the operation and control were located in a neighbourhood about 30 km away.
The treatment units had Programmable Logic Controllers (PLCs) that performed the interface between sensors and actuators. The station reported its status by dial-up to the monitoring center in the event of a predetermined event, alarm, power failure, or programmed time.
Again, my main task was to assist in the construction of the graphical interface for the supervision and control of the units, as well as to translate technological issues for the public, write manuals, and informational documents.
The operating interface was written in C using the NCurses Library. In addition, a non-interactive interface in an HTML browser provided monitoring information for technicians.
About HYDRA UX/UI design
The graphical representation was very literal and tried to be as faithful as possible to the functioning diagram, so that the operator could clearly understand what was happening.
The company's website, in which I also participated in development, showed performance graphs and water analysis updated once a day for end-users.
Currently, I think that UX research before the start of the project and usability tests should be done, in addition to responsiveness (nowadays, it is easy to imagine monitoring and control carried out through a tablet or smartphone).
TĪMA interface
A Berlin-based start-up called mergedK has developed a product for the electrical sector called Tīma GPS Master Clock. The equipment uses the time signal provided by GPS satellites to provide a time reference for the interconnected electrical system.
I participated in the development of the web interface for equipment configuration. I was also responsible for the visual identity, reflected in the design of the front panel, website, manuals, and promotional material.
Additionally, I wrote internal use software for automation of adhesive label generation, reflecting the varied configurations (due to made-to-order) for each model.
The Master Clock has an embedded Linux-based system and can be configured by uploading a text file. But, to make parametrisation more accessible to installation technicians, we developed a web interface that used Flask as a server and parameters could be changed on a screen generated with HTML and Javascript. Real-time monitoring used the MQTT protocol.
About TĪMA UX/UI design
Once again, the entire screen design took into consideration mainly information architecture, grouping similar functions and opening screens according to parameter details.
No usability research was conducted, although user feedback was spontaneously positive.
I'm sure that more in-depth UX Design knowledge would have contributed significantly to this development, especially in testing and research. Visual prototyping tools like Figma would have been really valuable.
MORe UX Design
In 2021, a Brazilian University hired me to teach a UX Design course in a postgraduate discipline. I did what I always do in these cases; conducted extensive research, bought and studied several books, compiled and translated everything in a structured and friendly way.
In 6 hours of course, I was able to build an overview of the main concepts and ideas and reach the current state-of-the-art.
The course was very well received, to the point that another University hired me for a more in-depth module; this time with more technical information.
CONCLUSIONS
Despite only now deciding to change careers, as you can see, I've been working with interfaces and have been interested in the subject for many years.
The knowledge acquired over the years will certainly be of great value for future projects, where the most appropriate methods and tools can be applied.