Racing Robocars
summary
A global DIY Robocar community of about 10,000 people across 40–50 regions builds wheel‑to‑wheel autonomous racers using cameras, Raspberry Pi or Jetson Nano boards, TensorFlow and cloud training, then run models locally. Typical approaches are behavioral cloning and reinforcement learning; platforms include Donkey Car and AWS DeepRacer and hardware like the Intel T265 visual odometry and $89 lidar. Race times drop from about six seconds per lap for cars toward a likely human best of ~4.5s; events run quarterly and often weekly around the world.
AI-written summary · about
chapters (5)
- 0:00Start
- 11:30Aggressive Driving
- 13:03Donkey Card Simulator
- 13:43Computer Vision
- 14:33Minimum Viable Racer
related talks
49mBuilding a self driving RC car
37mAWS DeepRacer: Train, Evaluate, and Tune
37mAWS DeepRacer:train, Evaluate, and Tune Your Reinforcement Learning Model
01h 51mLearn to Drive (and Race!) Autonomous Vehicles
35mDingoCar: Perception module Small scale, self-driving car
15mFrom the Racetrack to the Road - How Open Source Boosts Autonomous Vehicle Development
40mBuilding an Autonomous Robotic Vehicle
05mInterview with