Algorithms for Sensor-Based Robotics:

Introduction and Background

Computer Science 601.463/663

https://cirl.lcsr.jhu.edu/SensorBasedRobotics/
Simon Leonard

With slides borrowed from Zach Dodds and Howie Choset

EN.601.463/663 Simon Leonard

 Robotics from 50,000 feet
Task level planning

Haptics Mapping and Localization HRI

Dynamics and Control

(ME 646)

Mechanisms

Motion Planning
Vision Manipulation

Applications
EN.601.463/663 Simon Leonard
 Two Key Problems

1. Motion Planning: Given a mechanism, a known

environment, and a known start and goal, compute a
set of inputs (e.g. a joint trajectory) that moves the
robot from start to goal without collision.

2. Localization and Mapping: Given no a priori

knowledge, use information from sensing and motion
to simultaneously compute a model of the
environment (a map) and robot location within the
environment.

EN.601.463/663 Simon Leonard

 What is Motion Planning?

• Determining where to go
...more than a search (or it is a geometric search)

EN.601.463/663 Simon Leonard

 Basic Path Planning

Problem Statement:
Compute a continuous sequence of
collision-free robot configurations 2D EXAMPLE:
connecting the initial and goal
configurations

1. Geometry of
environment
Path
2. Geometry and Collision-free path
kinematics of robot Planner
3. Initial and goal
configurations
EN.601.463/663 Simon Leonard
 What did we assume?

• Perfect sensors?
– What information
– Uncertainty
• Perfect control?
– What controls?
– Uncertainty
• Perfect thinking?
– Knowledge of the world? Complete?
– Processing the world? Everything?

EN.601.463/663 Simon Leonard

Probabilistic Methods/Drug Design

Courtesy Lydia Kavraki et al.

EN.601.463/663 Simon Leonard
Motion Planning with Kinematics Constraints

Tesla Motors Otago Polytechnic

EN.601.463/663 Simon Leonard

 SLAM
• Suppose you wake up in the desert and you need to find your
way around --- how do you do it?

• A Chicken and Egg problem?

– How can you localize yourself without a map?
– How can you create a map without localization?

EN.601.463/663 Simon Leonard

EN.601.463/663 Simon Leonard
EN.601.463/663 Simon Leonard
EN.601.463/663 Simon Leonard
EN.601.463/663 Simon Leonard
EN.601.463/663 Simon Leonard
EN.601.463/663 Simon Leonard
Localization

EN.601.463/663 Simon Leonard

 This Course:
• We will focus on two robotics problems:
– Robot Motion Planning: How do I get from A to B in a known
environment
– Robot Localization and Mapping: What is the structure of space, and
where am I in it.

• We will primarily focus on algorithms, their analysis, and their

implementation (when possible) from real sensor data

• We will focus on the *static* planning problem

– the world does not change
– our own momentum does not matter

EN.601.463/663 Simon Leonard

 Book
http://motionplanning.com

Principles of Robot Motion:

Theory, Algorithms, and
Implementations

H. Choset, K. M. Lynch, S.
Hutchinson, G. Kantor, W. Burgard,
L. E. Kavraki and S. Thrun,

MIT Press, Boston, 2005.

EN.601.463/663 Simon Leonard

 Other Books

Planning Algorithms Steven

Lavalle, Cambridge
University Prress, 2006

Free download:
Robot Motion Planning, http://planning.cs.uiuc.edu/
Jean-Claude Latombe, Kluwer, 1991.

An Introduction to
Probabilistic Robotics AI/Robotics
S. Thrun, W. Burgard, D. Fox Robin Murphy
MIT Press, 2006 MIT Press, 2006

EN.601.463/663 Simon Leonard

 Classical / Hierarchical Paradigm

Sense Plan Act

• 70s
• Focus on automated reasoning and knowledge representation
• STRIPS (Stanford Research Institute Problem Solver): Perfect world model, closed
world assumption
• Find boxes and move them to designated position

EN.601.463/663 Simon Leonard

 Shakey the robot

Stanford Research
Institute
• Radio
• TV Camera
• Sonar range finder
• Bump detectors

Robot Hall Fame induction

in 2004.
wikipedia

EN.601.463/663 Simon Leonard

 Stanford CART
1961
1961: Initially designed as tele-operated moon
rover by using a TV camera and a radio link
1963: Added analog computer to improve
controllability
1971: Able to follow a white line at 0.8 mph
1980: Navigate around obstacles
Lester Earnest

1973

Lester Earnest
frc.ri.cmu.edu
EN.601.463/663 Simon Leonard
 Historical Search-Based AI
Blocks World (1960s)

EN.601.463/663 Simon Leonard

 Historical Search-based AI
STRIPS Action Planning (1960s)

EN.601.463/663 Simon Leonard

 Reactive / Behavior-based Paradigm

Sense Act

• No models: The world is its own, best

model
• Easy successes, but also limitations
• Investigate biological systems

EN.601.463/663 Simon Leonard

 Reactive Robots

groups.csail.mit.edu
EN.601.463/663 Simon Leonard
Reactive Robot

EN.601.463/663 Simon Leonard

 Trends in Robotics Research
Classical Robotics (mid-70’s)
• exact models
• no sensing necessary

Reactive Paradigm (mid-80’s)

• no models
• relies heavily on good sensing

Hybrids (since 90’s)

• model-based at higher levels
• reactive at lower levels

Probabilistic Robotics (since mid-90’s)

• seamless integration of models and sensing
• inaccurate models, inaccurate sensors

EN.601.463/663 Simon Leonard

Robotics Today

EN.601.463/663 Simon Leonard

 Robots in Space

CSA DARPA
Dextre Orbital Express

EN.601.463/663 Simon Leonard

 In the Air

Lily Camera

Air Force

EN.601.463/663 Simon Leonard

Land and Sea ...

EN.601.463/663 Simon Leonard

Intuitive Surgical da Vinci System

EN.601.463/663 Simon Leonard

 Robotics Sales
• Service Robots (two or more axis)
– Professional Service Robots: US$3,77
billion in sales in 2015
• Used for commercial tasks: surgical,
delivery, cleaning, military, …
– Personal Service Robots: US$2,2 billion
in sales in 2014
• Used for cleaning, mobility, exercise,…
• Industrial Robots (3 or more axis)
– US$10.7 billion in sales in 2014
– US$32 billion if considering software
and peripherals
Source www.ifr.org

EN.601.463/663 Simon Leonard

• 1 Midterm (25%), 1 Final (25%)
Spring 2018
• 5 assignments (10% each)
– 3 assignments with UR-5 in Wyman 170 (1% per
assignment demo)
– 2 assignment with Husky (1% per demo)
– Demos will be in teams of 2 or 3 (depending on
enrollment)
– Limited use of robots
• Use Piazza for questions, requests, etc. And
Blackboard to access and submit assignments
Assignment Assignment Assignment Assignment Assignment
#1 #2 #3 #4 #5

ROS
Gazebo
C++
Matlab
Demo UR5 EN.601.463/663
UR5 UR5 Jackal Simon Leonard
Jackal
 Spring 2018

• Ressources:
– Wyman 170:
• 6 UR5 + 1 desktop
• 3 additional desktops
• 2 Jackals
– Wyman 154:
• 14 desktops

• No curving
• No senior
A+ (97%-100%)
option A (94%-97%) A- (90%-94%)
B+ (87%-90%) B (84%-87%) B- (80%-84%)
C+ (77%-80%) C (74%-77%) C- (70%-74%)
D+ (67%-70%) D (60%-67%)

EN.601.463/663 Simon Leonard