This project aims at detecting and tracking moving objects with a 2D laser scanner independent of their classes and shapes. In this work, a Bayesian framework is proposed where the observations in our model are raw laser measurements.
The video to the right shows the system performance during a drive through central Oxford. Detected objects are marked with bounding boxes. Green boxes represent true detections and red boxes represent false alarms. Velocity estimates are shown with arrows. An image stream from an onboard camera is also shown to provide some context for visualisation. However our system does not require vision data to operate.