McLeod Software launches virtual AI Truck Racing League

McLeod Software is seeking teams to compete in its inaugural virtual AI Truck Racing League. Photo: McLeod

McLeod Software launched a virtual Artificial Intelligence Truck Racing League (AITRL) on Sept. 16, which is designed to educate the transportation industry about the importance of machine learning through gamification.

So far, eight companies have signed up to participate in the inaugural AITRL competition. Teams in the league will compete in time trials using algorithms to teach their 1/18 scale semi-trucks to race on virtual tracks in a 3D simulator inspired by famous raceways.

“Some trucking companies that are participating said they are concerned that their truck drivers will look at this competition as making an effort to get into autonomous trucking and get rid of the drivers, but this isn’t the case at all,” Ken Craig, vice president of special projects at McLeod, told FreightWaves.

The goal of the competition is to educate the industry about machine learning because the technology is becoming increasingly more valuable in helping companies speed up their accounting and back-office function processes, Craig said.


After the league is finalized, teams will have a few months to train their virtual trucks using cutting-edge AI technologies before structured races start.

“As the competition progresses, different tracks will be used during the timed trials,” Andrew Boyd, senior developer of McLeod, told FreightWaves.

Boyd, who also serves as the chief technology officer of AITRL, said different tracks will be used to prevent “overfitting” in machine learning, where a team trains its virtual truck to only perform well on one track but isn’t able to teach it to adapt to different virtual track conditions.

The plan is to extend this virtual racing opportunity to local high school students to “help them learn more about artificial intelligence and machine learning techniques,” Craig said.


Champions of the timed trials will compete for the AITRL League Cup, which is tentatively scheduled for December or early January 2020, Craig said.

“Machine learning is not yet widely understood by many, but to help companies gain a better understanding, we have set up a fun racing simulation as a way for companies to learn and participate,” he said.

Exit mobile version