Welcome to Zoom lectures: T/Th 10:00 11:20 (recordings on Canvas) - - PowerPoint PPT Presentation

welcome to
SMART_READER_LITE
LIVE PREVIEW

Welcome to Zoom lectures: T/Th 10:00 11:20 (recordings on Canvas) - - PowerPoint PPT Presentation

Organization Welcome to Zoom lectures: T/Th 10:00 11:20 (recordings on Canvas) Zoom office hours CSE 571 Robotics Dieter: Fri 9am Chris: Mon 4pm Xiangyun: Wed 2pm Tasks 4 homeworks covering Gaussian processes,


slide-1
SLIDE 1

1

SA-1

Welcome to CSE 571 Robotics

Instructor Dieter Fox Teaching Assistants Xiangyun Meng Chris Xie

  • 1

CSE-571: Robotics 3/31/20 2

Organization

§ Zoom lectures: T/Th 10:00 – 11:20 (recordings on Canvas) § Zoom office hours

§ Dieter: Fri 9am § Chris: Mon 4pm § Xiangyun: Wed 2pm

§ Tasks

§ 4 homeworks covering Gaussian processes, particle filters, RRT planning, and deep learning § Team project on simulation platform of your choice

§ Readings: Papers and chapters from Probabilistic Robotics § Web page: http://www.cs.washington.edu/571

  • 2

CSE-571: Robotics 3/31/20 3

Actions Control system Sensor data World model

High-level View on Robot Systems

  • 3

CSE-571: Robotics 3/31/20 4

Industrial Robotics Today

  • 4
slide-2
SLIDE 2

2

CSE-571: Robotics 3/31/20 5

Minerva

(CMU + Univ. Bonn, 1998)

  • 5

CSE-571: Robotics 3/31/20 6

Architecture of the Control System

  • 6

RoboCup: Integrated System Research

  • Focus on addressing all problems at once
  • Hardware development
  • Perception
  • Low level control
  • High level planning and decision making
  • Multi robot systems

CSE-571: Robotics 3/31/20 7

  • 7

CSE-571: Robotics 3/31/20 8

RoboCup-99, Stockholm, Sweden

  • 8
slide-3
SLIDE 3

3 RoboCup: Standard Platform

CSE-571: Robotics 3/31/20 9

  • 9

DARPA Urban Challenge 2007

CSE-571: Robotics 3/31/20 10

  • 10

Self-Driving Cars

CSE 571: Robotics 11

  • 11

Robots in Warehouses

(Kiva@Amazon)

CSE 571: Robotics 12

  • 12
slide-4
SLIDE 4

4 Amazon Prime Air

CSE 571: Robotics 13

  • 13

DARPA Robotics Challenge 2015

CSE-571: Robotics 3/31/20 14

  • 14

Getting out of Car

CSE-571: Robotics 3/31/20 15

  • 15

Drilling Hole

CSE-571: Robotics 3/31/20 16

  • 16
slide-5
SLIDE 5

5

Humanoid robots

CSE 571: Robotics 17

  • 17

Boston Dynamics BigDog

CSE-571: Robotics 3/31/20 18

  • 18

Boston Dynamics Spot

CSE-571: Robotics 3/31/20 19

  • 19

Boston Dynamics Atlas

CSE-571: Robotics 3/31/20 20

  • 20
slide-6
SLIDE 6

6 Boston Dynamics Handle

CSE-571: Robotics 3/31/20 21

  • 21

Industrial Pick and Place

CSE-571: Robotics 3/31/20 22

  • 22

Manipulation

CSE 571: Robotics 23

  • 23

Service Robots

CSE 571: Robotics 24

  • 24
slide-7
SLIDE 7

7 Dexterous Manipulation

CSE-571: Robotics 3/31/20 25

  • 25

HaptX Dataglove

CSE-571: Robotics 3/31/20 26

  • 26

Simulation

CSE-571: Robotics 3/31/20 27

  • 27

Current Trends / Topics

  • Self-driving cars, sidewalk delivery robots,

warehouses, manufacturing sites, …

  • Drones
  • Industrial pick and place
  • Manipulation of everyday objects
  • Complex household tasks (cooking, cleaning, …)
  • Object detection, 3D mapping, tracking, interaction
  • Cobots, human robot interaction
  • Deep learning for perception, control, imitation

learning, recognition

CSE-571: Robotics 3/31/20 28

  • 28
slide-8
SLIDE 8

8

CSE-571: Robotics 3/31/20 29

Goal of this course

  • Provide an overview of fundamental

problems / techniques in robotics

  • Understanding of estimation and decision

making in dynamical systems

  • Probabilistic modeling and filtering
  • Deterministic and non-deterministic planning
  • Learning for perception and modeling
  • 29

30

Course Outline

Week Content

HW / Project

#1 Introduction / Probabilities

Probabilistic Models / State Estimation

#2 Gaussian processes, Bayesian filtering HW1 assigned #2 Motion and sensor models

Filtering (localization, tracking, mapping)

#3 Localization: grid, particle filters, EKF , UKF #4 / 5 Mapping: SLAM, RGBD 3D Mapping HW2 assigned

Planning / Control

#6 / 7 Deterministic and sampling-based planning, exploration HW3 assigned #8 Markov decision processes, inverse RL

Deep Learning

#9 Model learning, visual navigation HW4 assigned #10 Grasping 3/31/20 CSE-571: Robotics

  • 30