The story of Wallace, our office robot. Part 1

The story of Wallace, our office robot. Part 1


Here at Jayway, competence development has always been a crucial part of our culture. Apart from founding, hosting and attending one of the biggest developer conferences in Europe – Øredev, we also enjoy internal conferences abroad and at our offices several times per year.
Last year we transformed into smaller software studios, each with the capabilities to take on any project.

Apart from having company-wide competence events we decided to have some ongoing competence projects going on in the studios.
This is the first part in a series on the studio named Braveheart and their quest to build a robot, aptly named Wallace.

The project

The studio project should accomplish three goals:
– Let the studio members evaluate and learn new technologies while spreading that knowledge in the studio and within the company
– Contribute to the open source community
– Have something cool to show off when people visit our offices

So in the spring of 2017, as we were at a competence retreat outside of Barcelona, Spain, we gathered to decide on a studio project. There are quite a few disciplines represented in each studio; iOS, Android, Backend, Web, Design, Project Management. We needed to think of a project that covered all of the three goals and also had challenges that suited most of the studio members. As all studio members have their core competencies, they also have areas of interest that they wish to further develop.

After a voting session we all decided that some form of robot would be a really cool thing to work on. The discussions of what the robot future capabilities quickly got out of hand so we needed to narrow the scope in order to have something that would take no more than a year to complete.

Long term plan

In the long term we figured that a robot can increasingly evolve with the times. The hardware is reasonably cheap and it is very software-driven.

Our hope for this project is to have a robot that can navigate our offices via voice commands (`Hey Wallace, go to the door and greet the visitors`) and also assist visitors with finding the way and contacting the employees they are here to see.

There are quite a few technical challenges right there but all of them are solvable and we really looked forward to solving them!

– How will the robot know its location and the office floor plan?
– How will the robot speak and react to spoken commands?
– How will the robot recognize people?
– How will the robot know that a visitor has arrived?
– How will the robot move?
– How will the robot power itself?

One thing was for sure, we had a lot of work to do and it would require the expertise and passion for development from all of us.

As a starting point it was decided that the “brains” of the robot would be a Raspberry Pi running Android Things. There were a few reasons for this.
For one, it’s simple to get started with and one gets a lot of the built in Android functionality for free. Also, we could write everything in Kotlin!
The Raspberry Pi allows for connecting other devices/peripherals using USB or Serial

So! We had a lot of work to do if we were going to get this done in under a year.

Stay tuned for part 2 where we go through how we went from idea to v1.0:


Leave a Reply