Taking the TT track at 7 km/h
Racing without a driver
It’s Wednesday morning and things are not going smoothly at the TT Junior Track in Assen. The kart refused to move: there’s an issue with the Wi-Fi. Team members Emma Langeveld and Ginnie Renz are frantically trying to figure out what the problem is.
Over the walkie talkie, they ask a staff member with the Dienst Wegverkeer (RDW) for help. After rushing over on his electric scooter, however, he quickly realises he can’t help them.
Fortunately, the students quickly find the problem: there’s a spelling error in the code, preventing the laptop from connecting to the kart.
How about now? Alas, it’s not working just yet. ‘Have you tried turning it off and on again?’ Emma asks Ginnie, giggling. It turns out a folder was named wrong. After everything’s been sorted, it’s finally time for the UG’s self-driving kart to start moving.
First time
The team, captained by industrial engineering and management student Emma, isn’t the only team that will be trying to get a cart across the track today. Teams from the Hanze University of Applied Sciences, University of Twente, and the Rotterdam, The Hague, and Windesheim universities of applied sciences are also testing their cars. They’re all part of the Self Driving Challenge organised by the RDW, the government service in charge of driving licences and licence plates.
The goal is to make an electric driverless kart cross the track as quickly as possible
The goal is to make an electric driverless kart cross the track as quickly as possible, circumnavigating any and all obstacles placed on the track.
So how do you get a kart to drive by itself? The UG uses cameras and a feedback system. ‘Think of an electric toothbrush’, says Emma’s fellow student Judith Oeben. ‘A red light comes on if you push too hard on your teeth, which tells you to stop.’
The kart essentially works the same: when the cameras see a white line, the software tells the kart to go in a different direction.
Not feasible
The other five teams went for a different type of technology: artificial neural learning, inspired by the way people process information using the neurons in their brains. The UG team considered this technique, but ultimately rejected it. ‘It requires a lot of information and we only had a short amount of time, so we didn’t think it was feasible’, says Judith.
In the future, the students would like to make the switch to LiDar (Light Detection And Ranging, or Laser Imaging Detection And Ranging). ‘It uses the same principle as radar’, Emma explains. ‘You send out a laser pulse which bounces back when it hits an object. The distance to the object is determined by how long it takes for the signal to return.’
They hope to use this advanced technology to not only avoid white lines, but also real obstacles on the road. While the kart currently still uses a feedback system, they’d like it to anticipate and be able to think in the future.
Women
The six-person team, put together by professor of mechatronics Bayu Jayawardhana, met once a week to prepare for the contest. They’re all STEM students from different disciplines, but only computing science student Rutger Wuijster knows how to code. Another interesting fact is that the UG team is the only one to have as many female members as male ones – the other teams have no women at all.
I’m finally seeing what I learned put into practice
While the other teams are participating in the contest as part of their internships, the UG students are doing this in addition to their studies. It takes up quite a lot of time – eight hours a week – and they don’t get paid for it. In the best-case scenario, if they do particularly well, they might end up with an internship, or perhaps a job at the RDW.
So what’s their motivation? ‘I’m finally seeing what I learned put into practice’, says Judith. ‘This challenge has shown me the applications, something I didn’t see as much at university. That alone makes all the work worth it.’
Low speed
With the sound of the Ferraris practising on the Grand Prix track for a race that will see them reaching speeds of more than two hundred kilometres per hour, the UG kart crawls across the Junior Track at seven kilometres an hour while the team cheers.
‘Our base code is working really well’, Judith says proudly. Applied mathematics students Ginnie explains why they’re so happy: ‘The RDW’s online simulation didn’t work, so we’re forced to test everything here on the track. We’ll work on expanding the code this afternoon.’
In theory, the kart could reach speeds of up to eighty kilometres an hour, but the team isn’t ready to try that yet. First, it has to be able to navigate properly at low speeds, since it’ll be easier to step in if something goes wrong. The karts, which were supplied by the RDW, aren’t just very sensitive, but they’re also very expensive, so the team doesn’t want it ending up on the grass.
The team does have some problems to figure out before the finale on June 8. During the test day, they realised their cameras occasionally mistake sunbeams for the white lines on the track. They also don’t yet know for sure if the track will have any extra obstacles. However, they have faith: they were the only team that actually got their kart to move today.