auto_car_interior
Article

Catch the checkered flag: highlights from the 2021 DeepRacer event

Racers, start your engines! And, jump into the driver’s seat to catch up on race highlights from our 2021 Baker Tilly Digital and AWS DeepRacer event.

DeepRacer is a virtual race league where participants drive a toy race car around the track through the use of artificial intelligence and machine learning commands. These cars use a type of machine learning called reinforcement learning, teams can make adjustments to the logic driving the car in order to improve the car’s performance and beat other race team’s track time. Anyone can join our next DeepRacer event, regardless of their technical experience, and you will have a Baker Tilly DeepRacer specialist assigned to your pit crew to chase down that checkered flag.

Let’s take a look at the highlight reel from Jan. 28, 2021’s race to see how the teams did on training their cars!


We greatly appreciate the team who built out a COVID friendly venue for the final so that we could virtually watch the cars with their trained models race around the track!

Each model was inserted into the car to use the unique capabilities each team harnessed so that the cars successfully move around the track. We have two models for this race. The first is named “JP Driving” and the second is named “Team Phil”. Each car will have only five laps to grab the best time! The moderator will increase the speed with each loop as the cars race around the track. Each time the car has all of the wheels off the track, the car is reset on the course.

  • The first model, JP Driving had a final best loop at 18.737 seconds.
  • The second model, Team Phil had a final time of 17.696 for their best lap. Taking first place for this championship!

We all look forward to how the teams will adjust their models in the future to grab that first place trophy!

Be the first to hear of the next event

arrowCreated with Sketch.
Banking KPI insights: 2017 second quarter metrics of note
Next up

Model Audit Rule program - reload your program and get ready to add value