CARLA, an open-source driving simulator used to support development, training, and validation of driving systems, announces the launch of the CARLA Autonomous Driving Challenge for participants to deploy state-of-the-art autonomous driving systems to tackle complex traffic scenarios.
Our world is shifting into an Autonomous Driving era, with the field of mobility going through a revolution that’s changing how we understand transportation.
Although the benefits of autonomous vehicles include eliminating human-caused accidents, reducing carbon dioxide emissions and efficient energy use, there are many unanswered questions: What’s the best path toward fully autonomous driving? What’s the best combination of sensors for driving? Which components of driving should be hand-crafted? In which traffic situations do different algorithms fail?
There are no real answers to these critically important questions because evaluating these driving systems is extremely expensive.
The CARLA Autonomous Driving Challenge is a step toward democratizing autonomous driving research and development by countering high costs and expediting research.
In the challenge, participants will deploy state-of-the-art autonomous driving systems to tackle complex traffic scenarios in CARLA. Referred to as “agents,” autonomous driving systems will demonstrate their proficiency by driving along complex urban and highway scenarios. The organizers selected traffic situations from the NHTSA pre-crash typology, which include negotiations at traffic junctions, dealing with pedestrians, lane merges and more.
The challenge will evaluate how proficiently and safely agents address each situation – and consists of four parallel tracks, focusing on different possible configurations of AVs. For each track, three winners will be awarded monetary prizes. Teams from both academia and industry are encouraged to participate.
The challenge is generously sponsored by industry leaders AWS, Waymo, Uber ATG, Audi EV and AlphaDrive – all of whom support advancements in autonomous driving.
The CARLA AD challenge will help to expedite the race towards AV-based mobility. The challenge results will allow for a better understanding of the benefits and problems of current top approaches for developing AVs, shaping the focus of future research efforts – and, in turn, determining the new aspects that must be assessed.
For further information, please visit https://carlachallenge.org.