From the Racetrack to the Road - How Open Source Boosts Autonomous Vehicle Development
summary
A Technical University of Munich race-team builds a full autonomous software stack that runs on Indy and Las Vegas speedways, achieves overtakes and peak speeds of 270 kph, and is being released as open source. Hardware uses GPS LiDAR camera and radar; perception runs three parallel detectors plus tracking; localization fuses GPS and LiDAR; planning includes safety checks and a model predictive controller. Middleware and deployment use ROS 2 Galactic, Eclipse Cyclone DDS and Docker, and a radar reflection once caused a high-speed spin.
AI-written summary · about
chapters (8)
- 0:00Research journey and origin
- 0:54Autonomous racing experiments
- 3:14Indie Autonomous Challenge
- 6:11Autonomous system stack
- 7:20Role of open source tools
- 8:40Future vision for Drive Blocks
- 10:12Building an open ecosystem
- 13:21Benefits of open source
related talks
49mBuilding a self driving RC car
23mRacing Robocars
27mPreparing for the Autonomous Vehicle Era
22mSteering into Tomorrow: Embracing Software-Defined Vehicle in an Open-Source Appro...
01h 51mLearn to Drive (and Race!) Autonomous Vehicles
19mDriving Automotive with Open Source: From Compliance to Collaboration
40mOpen Source in Automotive BoF
39mPanel Discussion: Open Source as a Path for a Competitive Automotive Industry