Formula Trinity is the College’s own student-led entry to Formula Student (FS), an engineering competition. The project was first started in 2017, but it can take teams many years of competing at various levels before they have a physical car to enter into the Class 1 dynamic racing events. In the years since its founding, the team has gone from strength to strength. They entered their first designs in the 2017/2018 Class 2 FS competitions. In their second year, they saw great success jumping to second place, from twenty-second the previous competition.
The team were gearing up to move to the Class 1 competition and compete with a physical car last year, but naturally, in the wake of Covid-19, everything became completely virtual. This year they continue to work on their internal combustion car and will enter it in the FS competitions in July. Aside from making the jump to the next level with the combustion engine team, Formula Trinity has another exciting project in the works this year. The group has set up an autonomous driving project, with the eventual aim of building a self-driving, AI-controlled car.
The autonomous project originally began in 2018, but fizzled out. This new iteration and relaunch of the autonomous driving team was the brain-child of members of the electronics team for the combustion engine team, now co-leads Andrew Dai, Jakub Pyszka and Senan Stanley. They are also joined by Katherine Hardgrave, operations co-lead. The autonomous project now has over 25 members, and they’ve made considerable progress in the past year.
Hardgrave, a third-year mechanical and manufacturing engineering student, attributes the foundation of the team to experiences co-lead Andrew Dai had on an exchange at the National University of Singapore: “Andrew Dai was over at NUS in Singapore and when he came back he heard about this competition, F1tenth.” Hardgrave explains that NUS is “huge into robotics. So I think he got a lot of inspiration there, and it really sparked his love for all this kind of stuff.”
The premise of F1tenth is that “basically you build an autonomous driving car, which is a tenth the size of a Formula One car. That’s kind of their shtick. The competitions for that are held worldwide. Last year’s one was supposed to be in Vegas. But obviously with Covid and everything that was just an entirely virtual competition.” The competition began as a module in the University of Pennsylvania and has since been expanded to a global initiative with all its resources available open-source, for the various teams that would like to take part.
Hardgrave started off as a member of the business and operations team for the internal combustion project, but jumped ship and became the lead of autonomous business operations. The work of the two teams isn’t currently interlinked much, as they are at very different stages of the two projects, with different objectives: “It’s under the same kind of Formula Trinity umbrella, but we’re pretty separate in the ways we operate. They’re two separate projects, the autonomous driving project and the internal combustion project. You don’t end up working together because they’re focused on very different things.”
Where the internal combustion team is focused on the Formula Student competitions and preparing their car, the internal combustion team is setting up a competition of their own. “The F1tenth competition, they normally hold two a year, one in April and one in like October, but it’s not being held this year in April, so we decided to host our own,” says Hardgrave. “That’s what we’re working on at the moment, to host our own virtual racing competition for our members to compete in. And also, we’re inviting other colleges in Ireland and other people in Trinity that aren’t on the team to take part as well.” The competition is set to take place on May 29th, and the team says it is a great way for Trinity students who are considering applying to be on the team next year to get a feel for the project.
The autonomous driving team is made of five research departments, similar to the original combustion engine team. David Nugent, a third-year computer science student, is the lead of one such committee: state estimation. Nugent points out that while Formula student teams themselves are fairly rare in Ireland, the autonomous team is the only one of its kind in the country: “It’s definitely the only autonomous driving team in Ireland, but also, there are actually not many organisations, in universities in Ireland that are doing anything to do with AI. I’m a computer science student and I’m really into AI. So I never joined formula training, because I wasn’t a huge F1 fan. But then this year, when I saw that they were doing autonomous driving that was like, the perfect fit because it’s AI as well. It’s hard to find places to work on AI, even though it’s starting to become hugely popular in universities already.”
On the perception that Formula Trinity is just for engineering students, Hardgrave says that the autonomous team will help to change that.
“We’re so new, we’re really just setting up and establishing the team now this year. So we haven’t got the proper recognition for all the stuff that our team’s doing on the autonomy side of things. So we’re kind of trying to get our name out there because what we’re doing is really cool!” The autonomous driving team has a lot more scope for wider student engagement from courses like computer science and from students with coding skills, whereas the combustion team needs a certain level of mechanical engineering skills and knowledge.
Working on the autonomous project is actually a great way to get valuable AI-based experience notes Hardgrave. “All the current concepts in AI, they’re not really taught at undergrad level, they’re taught to masters students. Especially in computer engineering, you won’t go near them until then, you might get to a few in computer science.” “It’s something where everyone wants to get into the industry. And it’s so in demand to have actual skills working on projects with this. It’s a huge opportunity for people to get some experience. There are so many resources out there, that if you have people around and you are learning it with them and helping each other out, you’ll learn a lot more.”
Nugent comments that he was actually looking for work in self-driving companies when he found out about the autonomous project, so it was “a really good fit. It’s impossible to get the experience for that in university, the same way it’s impossible to get racing engineering experience at university. That’s why a lot of the f1 guys would join our internal combustion team because that’s the only way you can get experience.” He also says that the social aspect of the team has been crucial this year and that with the number of meetings and amount of contact they have each week, he barely notices the lack of in-person interaction.
“But outside of that, of the social aspect, it’s really good to finally meet some people who actually had similar interests in AI and stuff, because you actually learned loads just by all the resources we share with each other. As well, it’s hard to find places in Trinity where you can actually work on a development project in computer science in general, where we can code up a big project together, that’s going to be a long term, existing solution. You get to learn how to create something in a big organisation, it’s really good experience for that as well.”
Remote working hasn’t been such a barrier to the autonomous team due to the stage of their project. “We wouldn’t be able to do the hardware stuff anyway, because we’re so new we would just still be doing kind of virtual online stuff anyway. So it hasn’t hindered our progression too much yet,” explains Hardgrave. They spent the first half of this year training the team, and the three other co-leads designed a crash course on Github to train in their new members.
Now later in the year, the research departments have been formed and the teams meet regularly to work on their various responsibilities in the project. Nugent runs coding labs and has created youtube tutorials to assist the team in their learning. He says that it is of great benefit for them to code together “We just try to make everything really streamlined and easy for everyone. Because it’s hard to motivate everybody, especially right now. And you can’t really keep track of everybody and everyone’s progress because you don’t get to see them. So we just try to make everything as easy as possible for people.”
But aside from working towards making a driverless car and coordinating their own virtual autonomous driving competition, they have also coordinated a campaign this month. In February Formula Trinity ran a month-long campaign to raise awareness of women in motorsports. Formula Trinity has just begun a new partnership with Women in AI Ireland. “Formula Trinity is highly passionate about not only recruiting more women onto the team but highlighting the work & research done by women in the industry too,” they say.
Now Hardgrave is setting up a new campaign: AI-pril for this month. The hope is to promote the new autonomous team, with the campaign including talks with industry leaders as well as a podcast “Formula Trinity Talks”, the most recent featuring founder and CEO of Manna Aero, a drone delivery service. They are holding a panel discussion on April 28 in collaboration with their new partner, Women in AI Ireland, to “highlight the opportunities in the fields of robotics, autonomous vehicles and data science.”
In the next few years, the goal for the autonomous team is to have a physical self-driving car to race at the F1tenth and Formula Student AI competitions. They are hoping to be able to retrofit the combustion teams car with sensors for this purpose, which is a fitting recombination of the two separate teams with the same roots. Hardgrave says one of the strengths of the team and her three co-leads is their focus on the future. “The three guys have founded it, Jakub, Senan and Andrew, they’re so passionate about the future of the team. Andrew and Senan are both graduating this year and they’ve done so much work to make sure that the team will continue on. We want to be really certain that there were enough people and enough young students that would stay on for the next three years and keep the team going.”
For anyone interested in taking part in their competition on May 29, or applying to be on the autonomous team next year, Hardgrave and Nugent say the most important attribute a team member can have is dedication. “We’re not looking for a coding prodigy that has so many other projects, they won’t be able to give the time to this. They could be amazing, but not gonna put in the time, the hard work. We have people on the team that are really willing to learn and genuinely want to put in the work.”