IntentNet: Learning to Predict Intention from Raw Sensor Data
This addresses the need for safe maneuver planning in autonomous driving, though it is incremental as it builds on existing detection and forecasting methods.
The paper tackles the problem of predicting the intent of other traffic participants for self-driving vehicles by combining discrete high-level behaviors and continuous trajectories from raw sensor data, achieving better accuracy than separate modules while saving computation.
In order to plan a safe maneuver, self-driving vehicles need to understand the intent of other traffic participants. We define intent as a combination of discrete high-level behaviors as well as continuous trajectories describing future motion. In this paper, we develop a one-stage detector and forecaster that exploits both 3D point clouds produced by a LiDAR sensor as well as dynamic maps of the environment. Our multi-task model achieves better accuracy than the respective separate modules while saving computation, which is critical to reducing reaction time in self-driving applications.