Team Interview #9< | >

Team: Watermelon Project


Country: Spain


Affiliation: Uni. of León

We recently spoke with Francisco Lera, Team Leader at Watermelon Project team, about the upcoming RoCKIn2015 event. Here's what he had to say about their mission, vision and preparation for the final event!

Hello Francisco - thank you for joining us today. To start with, could you give us a bit of background information on your team, its aims and mission?

Watermelon Project is a joint-project of two research groups from two Spanish Universities: Universidad de León and Universidad Rey Juan Carlos. We are a small team made up by just four people, which makes the coordination easier. All members of the team work autonomously and we only need consensus on high-level decisions.

Since the beginning, our goal has been the development of a low-cost, replicable robotic named MYRABot for research environments. From the @Home point of view, we focus our work on designing a robot that could be deployed by anyone in any home.

However, for the purposes of the competition, we are going to change our robot for this final RoCKIn event, using an RB-1 platform instead of the MYRABot platform. RB-1 is a more reliable and robust commercial platform manufactured by Robotnik SA, and this competition will be its "first" intensive test.

What appeals to you about RoCKIn and how does it complement your research and personal development goals? What about robot competitions more generally?

Robotic competitions define a set of deadlines that impose an external pressure that it is very important in research environments.

At the same time, robotic challenges generate a set of formal and strict milestones that you have to accomplish during the competitions days.

Finally, RoCKIn provides a formal measurement of our performances, as well as personalized feedback from the committee. This is an amazing help in presenting our contributions in a scientific manner.

Have you been involved in robotic competitions before? If so, what were some of the highlights for you?

Yes, some members of our team were involved in SPL RoboCup competition. That experience has provided the team with a solid background that helps us face the challenges associated to a robotic competition. For instance, we can tackle hardware problems due to travel restrictions, network issues due to massive number of participants, new surfaces that affects robot navigation or lighting conditions that can change the default parameters of our perception system.

What role do you think competitions have in furthering innovation in robotics?

RoCKIn, and other competitions, offers us an opportunity to meet, talk and discuss with other researchers about the different ways to address household tasks. At the same time, we do not have the restrictions of the lab and can compare and measure our performance in a more aseptic environment.

Scientific competitions challenge the community in several unique ways: firstly, it provides a common setup upon which the different teams, in equal terms, have to evaluate their solutions (and that is why the benchmarking aspect of RoCKIn is so relevant); second, it takes the teams out of the lab, by exposing them to real world conditions; thirdly, underlying competitiveness ends up pushing teams to their best capabilities; and finally, there is little tolerance for lack of robustness, thus forcing teams to explicitly address reliability issues. All of these aspects provide a big push for innovative solutions towards real-world problems.

How are your preparations for RoCKIn2015 going? What are you finding to be the most challenging areas?

Well, the competition is soon and we are working hard to arrive in Lisbon with the best version of our software and robot.

As a research group, the team members are interested in different topics such as perception, interaction, or navigation. We usually perform our individual research with one of these topics in mind. We think in how to accomplish a new contribution; consequently, we sometimes don't pay enough attention to the programming side and its impact on the overall system. This means that our approaches run perfectly in the robot as a standalone solution, but it is not always so easy to integrate it with the other parts of the system.

And finally, what do you think the future holds for domestic robotics?

The future is promising for domestic service robotics; however there is a long way to go before we see the platforms involved in the RoCKIn Challenge deployed in a real home.

Present-day robotics shows the vacuum cleaner robots as the kings and queens. They are one of the most successful platforms in the history of domestic service robotics. In our opinion, this is because they offer a balanced trade-off between tasks (only vacuuming), performance (they clean apartments and flats without problems), and price.

We should build on these premises and start thinking about how to translate our astonishing research work to the real world.

Thanks for your time Francisco - we're looking forward to seeing you in Lisbon!