Technical

Technical demonstrations of each teams progress.

Jetracer Optimized

Jetracer Autonomous DIY Optimized https://youtu.be/fgqukxg3LHc The Triton AI Jetracer team has cracked the code for the NVIDIA Jetracer DIY, optimizing how the Jetracer steering and throttle decisions made with PyTorch. The Triton AI Jetracer DIY robot now has the fastest lap time at 11.32 seconds because of the optimized training algorithm with carefully selected training

Jetracer Optimized Read More »

Jetsim

Jetsim is Alive! https://youtu.be/OslGWZIrPxMThe Triton AI Jetracer team has successfully built an autonomous driving model in the Donkeycar simulator using NVIDIA’s open source Jetracer framework.  We call this interface Jetsim! Jetracer was originally designed to operate on robotic race cars using the Jetson Nano developer kit. Triton AI is using this technology to interface with

Jetsim Read More »

Jetracer Pro Modification

Waveshare Jetracer Pro gets modified https://youtu.be/bwDbZgUgWmc Triton AI has added some modifications to the Waveshare Jetracer Pro kit! It is now controlled via radio controller making the car completely independent of the Jetson Nano until channel 3 is pushed on. This makes the car safer to train and emergency stop if needed. The Pro is

Jetracer Pro Modification Read More »

Jetbot

Jetbot https://youtu.be/-KbXeafMG50 Team Jetracer is experimenting with the Waveshare Jetbot to understand how Pytorch works in detail. Jetbot and Jetracer have a lot in common considering they are both written by Nvidia. In this video below, JetBot was trained on a neural network with a total of 80 images. 40 images were classified as a

Jetbot Read More »