Towards learning-based planning:The nuPlan benchmark for real-world autonomous driving
This work addresses the need for robust testing of ML-based planners in autonomous driving, though it is incremental as it focuses on benchmarking rather than a new planning method.
The authors tackled the slow adoption of machine learning in autonomous vehicle planning by introducing nuPlan, the first real-world dataset and benchmark for ML-based planners, which includes 1282 hours of driving data from 4 cities and enables closed-loop simulation to evaluate safety and efficiency.
Machine Learning (ML) has replaced traditional handcrafted methods for perception and prediction in autonomous vehicles. Yet for the equally important planning task, the adoption of ML-based techniques is slow. We present nuPlan, the world's first real-world autonomous driving dataset, and benchmark. The benchmark is designed to test the ability of ML-based planners to handle diverse driving situations and to make safe and efficient decisions. To that end, we introduce a new large-scale dataset that consists of 1282 hours of diverse driving scenarios from 4 cities (Las Vegas, Boston, Pittsburgh, and Singapore) and includes high-quality auto-labeled object tracks and traffic light data. We exhaustively mine and taxonomize common and rare driving scenarios which are used during evaluation to get fine-grained insights into the performance and characteristics of a planner. Beyond the dataset, we provide a simulation and evaluation framework that enables a planner's actions to be simulated in closed-loop to account for interactions with other traffic participants. We present a detailed analysis of numerous baselines and investigate gaps between ML-based and traditional methods. Find the nuPlan dataset and code at nuplan.org.