Lecture 11 - Particle Filters

Overview:

This is the third installment in the Localization lectures. Given inputs of laser scan data, a control input, and a map, we want to output a pose and some “particles”. This lecture is divided into three parts: 1) going over the particle filter localization algorithm, 2) tuning the alrogithm, and 3) additional resources and a hands on tutorial. Expect a dense but extremely informative lecture.

Topics Covered:

• Particle filters

Slides:

Links to additional resources:

• S. Thrun, W. Burgard. “Probabilistic Robotics.” Chapter 4 and Chapter 8.

• S. Thrun. “Artificial Intelligence for Robotics, Lesson 3.” Udacity.

• S. Thrun, D. Fox, W. Burgard and F. Dellaert. “Robust Monte Carlo Localization for Mobile Robots.”​ Artificial Intelligence Journal. 2001.

• D. Fox, W. Burgard, and S. Thrun. “Markov localization for mobile robots in dynamic environments,” ​ Journal of Artificial Intelligence Research ​ , vol. 11, pp. 391427, 1999.

• D. Fox. “KLD-sampling: Adaptive particle filters,” Advances in Neural Information Processing Systems 14 (NIPS), Cambridge, MA, 2002. MIT Press.

• D. Bagnell “Particle Filters: The Good, The Bad, The Ugly”

• C. Walsh, S. Karaman. “CDDT: Fast Approximate 2D Ray Casting for Accelerated Localization.” Arxiv, 2017.