© sebastian thrun, CMU, 20001 CS226 Statistical Techniques In Robotics Monte Carlo Localization Sebastian Thrun (Instructor) and Josh Bao (TA)

Slides:



Advertisements
Similar presentations
Probabilistic Techniques for Mobile Robot Navigation
Advertisements

State Estimation and Kalman Filtering CS B659 Spring 2013 Kris Hauser.
Sebastian Thrun Carnegie Mellon & Stanford Wolfram Burgard University of Freiburg and Dieter Fox University of Washington Probabilistic Algorithms for.
Probabilistic Robotics
Probabilistic Robotics SLAM. 2 Given: The robot’s controls Observations of nearby features Estimate: Map of features Path of the robot The SLAM Problem.
Monte Carlo Localization for Mobile Robots Karan M. Gupta 03/10/2004
IR Lab, 16th Oct 2007 Zeyn Saigol
Probabilistic Robotics
Markov Localization & Bayes Filtering 1 with Kalman Filters Discrete Filters Particle Filters Slides adapted from Thrun et al., Probabilistic Robotics.
1 Slides for the book: Probabilistic Robotics Authors: Sebastian Thrun Wolfram Burgard Dieter Fox Publisher: MIT Press, Web site for the book & more.
Bayesian Robot Programming & Probabilistic Robotics Pavel Petrovič Department of Applied Informatics, Faculty of Mathematics, Physics and Informatics
Recursive Bayes Filtering Advanced AI Wolfram Burgard.
Bayes Filters Pieter Abbeel UC Berkeley EECS Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics TexPoint fonts used in EMF. Read the.
Probabilistic Robotics Bayes Filter Implementations Particle filters.
Introduction to Mobile Robotics Bayes Filter Implementations Gaussian filters.
Centre for Autonomous Systems SWAR Sept 8, 2009 © 2009 Omid Aghazadeh Model Adaptation in Monte Carlo Localization Omid Aghazadeh.
Recursive Bayes Filtering Advanced AI
Localization David Johnson cs6370. Basic Problem Go from thisto this.
Probabilistic Robotics: Kalman Filters
Robotic Mapping: A Survey Sebastian Thrun, 2002 Presentation by David Black-Schaffer and Kristof Richmond.
Sebastian Thrun Carnegie Mellon University Statistical Learning in Robotics State-of-the-Art, Challenges and Opportunities.
CS 547: Sensing and Planning in Robotics Gaurav S. Sukhatme Computer Science Robotic Embedded Systems Laboratory University of Southern California
A brief Introduction to Particle Filters
Sebastian Thrun Carnegie Mellon University University of Pittsburgh Particle Filters In Robotics or: How the World Became To Be One Big Bayes Network.
Stanford CS223B Computer Vision, Winter 2005 Lecture 12: Filters / Motion Tracking Sebastian Thrun, Stanford Rick Szeliski, Microsoft Hendrik Dahlkamp.
Nonlinear and Non-Gaussian Estimation with A Focus on Particle Filters Prasanth Jeevan Mary Knox May 12, 2006.
Stanford CS223B Computer Vision, Winter 2006 Lecture 12 Filters / Motion Tracking 2 Professor Sebastian Thrun CAs: Dan Maynes-Aminzade, Mitul Saha, Greg.
Part 2 of 3: Bayesian Network and Dynamic Bayesian Network.
Robust Monte Carlo Localization for Mobile Robots
Monte Carlo Localization
Particle Filters for Mobile Robot Localization 11/24/2006 Aliakbar Gorji Roborics Instructor: Dr. Shiri Amirkabir University of Technology.
Probabilistic Robotics Introduction Probabilities Bayes rule Bayes filters.
A Probabilistic Approach to Collaborative Multi-robot Localization Dieter Fox, Wolfram Burgard, Hannes Kruppa, Sebastin Thrun Presented by Rajkumar Parthasarathy.
Probabilistic Robotics
© sebastian thrun, CMU, CS226 Statistical Techniques In Robotics Sebastian Thrun (Instructor) and Josh Bao (TA)
Probabilistic Robotics Bayes Filter Implementations Particle filters.
Stanford CS223B Computer Vision, Winter 2007 Lecture 12 Tracking Motion Professors Sebastian Thrun and Jana Košecká CAs: Vaibhav Vaish and David Stavens.
Stanford CS223B Computer Vision, Winter 2006 Lecture 11 Filters / Motion Tracking Professor Sebastian Thrun CAs: Dan Maynes-Aminzade, Mitul Saha, Greg.
Probabilistic Robotics Bayes Filter Implementations Gaussian filters.
Bayesian Filtering for Location Estimation D. Fox, J. Hightower, L. Liao, D. Schulz, and G. Borriello Presented by: Honggang Zhang.
© sebastian thrun, CMU, CS226 Statistical Techniques In Robotics Paper Style Guide Sebastian Thrun and Rahul Biswas
© sebastian thrun, CMU, CS226 Statistical Techniques In Robotics Sebastian Thrun (Instructor) and Josh Bao (TA)
Bayesian Filtering Dieter Fox Probabilistic Robotics Key idea: Explicit representation of uncertainty (using the calculus of probability theory) Perception.
HCI / CprE / ComS 575: Computational Perception
Bayesian Filtering for Robot Localization
Markov Localization & Bayes Filtering
From Bayesian Filtering to Particle Filters Dieter Fox University of Washington Joint work with W. Burgard, F. Dellaert, C. Kwok, S. Thrun.
Probabilistic Robotics: Monte Carlo Localization
Probabilistic Robotics Robot Localization. 2 Localization Given Map of the environment. Sequence of sensor measurements. Wanted Estimate of the robot’s.
Probabilistic Robotics Bayes Filter Implementations Gaussian filters.
1 Robot Environment Interaction Environment perception provides information about the environment’s state, and it tends to increase the robot’s knowledge.
Probabilistic Robotics Bayes Filter Implementations.
Mobile Robot Localization (ch. 7)
Processing Sequential Sensor Data The “John Krumm perspective” Thomas Plötz November 29 th, 2011.
City College of New York 1 Dr. Jizhong Xiao Department of Electrical Engineering City College of New York Advanced Mobile Robotics.
P ARTICLE F ILTER L OCALIZATION Mohammad Shahab Ahmad Salam AlRefai.
Simultaneous Localization and Mapping (SLAM). Localization Perfect Map + Observations with errors = Pretty good Localization (Average out errors in observations,
CSE-473 Project 2 Monte Carlo Localization. Localization as state estimation.
Probabilistic Robotics
HCI/ComS 575X: Computational Perception Instructor: Alexander Stoytchev
CS 547: Sensing and Planning in Robotics Gaurav S. Sukhatme Computer Science Robotic Embedded Systems Laboratory University of Southern California
Monte Carlo Localization for Mobile Robots Frank Dellaert 1, Dieter Fox 2, Wolfram Burgard 3, Sebastian Thrun 4 1 Georgia Institute of Technology 2 University.
Probabilistic Robotics Bayes Filter Implementations Gaussian filters.
Mobile Robotics. Fundamental Idea: Robot Pose 2D world (floor plan) 3 DOF Very simple model—the difficulty is in autonomy.
Probabilistic Robotics
Markov ó Kalman Filter Localization
State Estimation Probability, Bayes Filtering
Particle Filter/Monte Carlo Localization
A Short Introduction to the Bayes Filter and Related Models
EE-565: Mobile Robotics Non-Parametric Filters Module 2, Lecture 5
Presentation transcript:

© sebastian thrun, CMU, CS226 Statistical Techniques In Robotics Monte Carlo Localization Sebastian Thrun (Instructor) and Josh Bao (TA) Office: Gates 154, Office hours: Monday 1:30-3pm

© sebastian thrun, CMU, Bayes Filters Bayes [Kalman 60, Rabiner 85] x = state t = time m = map z = measurement u = control Markov

© sebastian thrun, CMU, Bayes filters  Linear Gaussian: Kalman filters (KFs, EKFs)  Discrete: Hidden Markov Models (HMMs)  With controls: Partially Observable Markov Decision Processes (POMDPs)  Fully observable with controls: Markov Decision Processes (MDPs)  With graph-structured model: Dynamic Bayes networks (DBNs)

© sebastian thrun, CMU, Markov Assumption  Past independent of future given current state  Violated: Unmodeled world state Inaccurate models p(x’|x,u), p(z|x) Approximation errors (e.g., grid, particles) Software variables (controls aren’t random)

© sebastian thrun, CMU, x t-1 utut p(x t |x t-1,u t ) Probabilistic Localization map m laser datap(z|x,m)

© sebastian thrun, CMU, What is the Right Representation? Multi-hypothesis [Weckesser et al. 98], [Jensfelt et al. 99] Particles [Kanazawa et al 95] [de Freitas 98] [Isard/Blake 98] [Doucet 98] Kalman filter [Schiele et al. 94], [Weiß et al. 94], [Borenstein 96], [Gutmann et al. 96, 98], [Arras 98] [Nourbakhsh et al. 95], [Simmons et al. 95], [Kaelbling et al. 96], [Burgard et al. 96], [Konolige et al. 99] Histograms (metric, topological)

© sebastian thrun, CMU, Particle Filters

© sebastian thrun, CMU, Particle Filter Represent p ( x t | d 0..t,m) by set of weighted particles {x (i) t,w (i) t } draw x (i) t  1 from p(x t-1 |d 0..t  1,m ) draw x (i) t from p ( x t | x (i) t  1,u t  1,m ) Importance factor for x (i) t :

© sebastian thrun, CMU, Monte Carlo Localization (MCL)

© sebastian thrun, CMU, Monte Carlo Localization (MCL)  Take i-th sample  “Guess” next pose  Calculate Importance Weights  Resample

© sebastian thrun, CMU, Monte Carlo Localization

© sebastian thrun, CMU, Sample Approximations

© sebastian thrun, CMU, Monte Carlo Localization, cont’d

© sebastian thrun, CMU, Performance Comparison Monte Carlo localizationMarkov localization (grids)

© sebastian thrun, CMU, What Can Go Wrong? Model limitations/false assumptions  Map false, robot outside map  Independence assumption in sensor measurement noise  Robot goes through wall  Presence of people  Kidnapped robot problem Approximation (Samples)  Small number of samples (eg, n=1) ignores measurements  Perfect sensors  Resampling without robot motion  Room full of chairs (discontinuities)

© sebastian thrun, CMU, Localization in Cluttered Environments

© sebastian thrun, CMU, Kidnapped Robot Problem

© sebastian thrun, CMU, Probabilistic Kinematics map m

© sebastian thrun, CMU, Pitfall: The World is not Markov!

© sebastian thrun, CMU, Error as Function of Sensor Noise sensor noise level (in %) error (in cm) 1,000 samples

© sebastian thrun, CMU, dual mixed MCL Error as Function of Sensor Noise sensor noise level (in %) error (in cm)

© sebastian thrun, CMU, Avoiding Collisions with Invisible Hazards Raw sensorsVirtual sensors added