probabilistic robotics by sebastian thrun

Techniques that enable camera-laser calibration online, automatically,Īnd in arbitrary environments. Of 3D laser sensors to 2D cameras, this paper presents two new real-time Robotics: Science and Systems (RSS), 2013.Įxtending previous work on offline 6-DOF calibration Group induction is a mathematical framework for this kind of learning.Īutomatic Online Calibration of Cameras and Lasers Qualitatively, this would mean the ability to point out - from the back seat of your autonomous car - a few examples of, say, an elliptical bike or sk8poler, and the algorithm would start learning to recognize them on the fly without you having to do anything else. This is fine in some contexts, but ideally a user could provide new hand-labeled training examples online, as the system runs, without retraining from scratch.

Tracking-based semi-supervised learning, as originally presented at RSS2011, was an offline algorithm. of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2013. Project Page - Supplementary material, C++ code, poster, presentation, We currently use this method to track all dynamic obstacles seen by our autonomous vehicle, in real-time, with significantly improved accuracy compared to our previous Kalman-filter based approach. The algorithm is “anytime”, allowing speed or accuracy to be optimized based on the needs of the application.

Now, computational time is allocated dynamically according to the shape of the track’s posterior distribution. Improving on our ICRA 2013 paper, this new approach enables real-time probabilistic object tracking. Robotics: Science and Systems (RSS), 2014. Combining 3D Shape, Color, and Motion for Robust Anytime Trackingĭavid Held, Jesse Levinson, Sebastian Thrun, Silvio Savarese.

Probabilistic robotics by sebastian thrun