The Robot Revolution has been Postponed (until we can debug the sensors) Bill Smart Oregon State University

Slides:



Advertisements
Similar presentations
Homework Assignment. Assignment One Ready for the next challenge. Great! Time to pick an affiliate product to promote. Go to ClickBank.com and go to the.
Advertisements

Sonar and Localization LMICSE Workshop June , 2005 Alma College.
Knowledge Engineering Week 3 Video 5. Knowledge Engineering  Where your model is created by a smart human being, rather than an exhaustive computer.
Hi_Lite Scott Fukuda Chad Kawakami. Background ► The DARPA Grand Challenge ► The Defense Advance Research Project Agency (DARPA) established a contest.
So you want to be a programmer!!!! Presented By: Frank Larkin Lansdale Catholic Robotics, Team 272 Philadelphia PA, Dec
1-1 Welcome to: CSC225 Introduction to Computer Organization Paul Hatalsky.
By Chris Sanyk. About Me First exposure to computers c Atari 2600, Commodore 64, Apple ][ Game concepts on paper at age years in various.
HCI Final Project Robust Real Time Face Detection Paul Viola, Michael Jones, Robust Real-Time Face Detetion, International Journal of Computer Vision,
Administrivia Any new students?
LAB 3 – Review of the Assignment. -- Clarifications Vikram Murali. TA : CSE 140L Prof. CK Cheng.
Feature Detection and Emotion Recognition Chris Matthews Advisor: Prof. Cotter.
Miscellaneous topics and advice Never ever ever ever ……… EVER ….. What you should never ever ever ever ever do Light bulbs, planters, tough experiemts,
Java Review 2 – Errors, Exceptions, Debugging Nelson Padua-Perez Chau-Wen Tseng Department of Computer Science University of Maryland, College Park.
Robot? What’s a Robot? Introducing Karel-the-Robot.
10-1 Programming Remember: –programming language –how to program (conceptually) –intro to programming the “ROBOT” computer In this lecture: –programming.
EG1003: Introduction to Engineering and Design Sensors.
Quick Overview of Robotics and Computer Vision. Computer Vision Agent Environment camera Light ?
Digital Citizenship By Bhavna. Plagiarism Plagiarism is illegal and can get you arrested. If a teacher finds out you used plagiarism he/she can fail you.
Robotics- Basic On/Off Control Considerations. On/Off Control Forms the basis of most robotics operations Is deceptively simple until the consequences.
Team Spot On! With Zebulon Clark Desiree And Earl.
Why do robots need to move?
Associative Pattern Memory (APM) Larry Werth July 14, 2007
INTERMEDIATE PROGRAMMING LESSON By: Droids Robotics Debugging Techniques.
The Nature of Classical Physics A Rehash Followed by Some New Stuff 1 Nature of Classical Physics TexPoint fonts used in EMF. Read the TexPoint manual.
Today’s Agenda 1.Scribbler Program Assignment 1.Project idea due next class 2.Program demonstration due Wednesday, June 3 2.Attendance & lab pair groupings.
Lecture 29: Face Detection Revisited CS4670 / 5670: Computer Vision Noah Snavely.
Robotics Light sensor. Calibration. Reverse engineering challenge. Lab work: Create and show your program. Add light sensor (different orientations). Robot.
Debugging Strategies from Software Carpentry. Agan's Rules Many people make debugging harder than it needs to be by: Using inadequate tools Not going.
1 CS Tutorial 5 Frid. Oct 23, 2009 Design Document Tutorial.
2008 SBPLI/FIRST Programming Workshop Tom Boehm Patchogue Medford High School, Team 329 Motorola Inc. Mark McLeod Hauppauge High School Team 358 Northrop.
Sentry System Multiple Sensors
Delivering Business Value through IT Face feature detection using Java and OpenCV 1.
Course Overview  What is AI?  What are the Major Challenges?  What are the Main Techniques?  Where are we failing, and why?  Step back and look at.
Have fun on the internet while being safe!!  Do you know what cyber safety means?.. YesNo.
David Wild Supervisor: James Connan Rhodes University Computer Science Department Eye Tracking Using A Simple Webcamera.
Lecture 10: Lines from Edges, Interest Points, Binary Operations CAP 5415: Computer Vision Fall 2006.
HOW THEY WORK AND WHAT THEY DO. Jay Jay 9S. A control program is a type of programming that allows you to control systems. Control programs are used in.
Fast SLAM Simultaneous Localization And Mapping using Particle Filter A geometric approach (as opposed to discretization approach)‏ Subhrajit Bhattacharya.
LEARNING LOGS AVID A learning log is something that helps you articulate what you’ve learned and discover what you don’t understand. It also helps.
1 RoboticsPresents KaaShiv InfoTech For Inplant Training / Internship, please download the "Inplant training registration form" from our website
Robot Project by Ahmad Shtaiyat Supervised by Dr. Salem Al-Agtash.
EG1003: Introduction to Engineering and Design Laboratory 4: Sensors.
CS 1114: Finding things Prof. Graeme Bailey (notes modified from Noah Snavely, Spring 2009)
Casne.ncl.ac.uk Taking care of the CrumbleBot Please do NOT stress the robot's motors 1.Do NOT push the robot 2.Do NOT hold the.
VISUAL storytelling VISUAL. 7 tips for great videos.
SLAM Techniques -Venkata satya jayanth Vuddagiri 1.
ROBOTC for VEX Online Professional Development. Homework Questions Thoughts? Questions?
Introduction to Robots and the Mind - Sensors - Bert Wachsmuth & Michael Vigorito Seton Hall University.
Cyber Info Gathering Techniques
Face Detection and Notification System Conclusion and Reference
VEX IQ Curriculum Smart Machines Lesson 09 Lesson Materials:
Exploring Computer Science Lesson 6-5
INTERMEDIATE PROGRAMMING LESSON
INTERMEDIATE PROGRAMMING Lesson
INTERMEDIATE programming LESSON: debugging Techniques: DISPLAY, LIGHT, Sound, Button Press… By Droids Robotics © 2014, Droids Robotics, v. 1.0,
Sensors For Robotics Robotics Academy All Rights Reserved.
ROBOTC for VEX On-Site Professional Development
CS b659: Intelligent Robotics
Copyright © 2008 by Helene G. Kershner
Copyright © 2008 by Helene G. Kershner
Sensors For Robotics Robotics Academy All Rights Reserved.
Networked Embedded Systems
Introduction to robotics
What Should I Say to a Suicidal Friend?
Algorithm and Ambiguity
In the land of the blind, the one eyed man is king
INTERMEDIATE PROGRAMMING LESSON
The value of the month November Honesty.
Elizabeth Liner Lab of Things.
Presentation transcript:

The Robot Revolution has been Postponed (until we can debug the sensors) Bill Smart Oregon State University

What’s Hard in Robotics?

Sense-Think-Act

Sense: Sensors

Sensors are... noisy wrong late

Sensors Measure the Wrong Thing hopefully its correlated with the right thing assumptions, assumptions, assumptions

The Sad Truth all sensors are terrible

Example Sensors LIDAR cameras kinect-like sensors

LIDAR

shoots laser light into the world measures return intensity what does it measure? noisy, wrong, and late

Cameras

detects light reflected from things in the world on an array of silicon what does it measure? noisy, wrong, and late

Cameras

Kinect-like Sensors

Stereo Vision

Kinect-like Sensors

[video]

Remember all sensors are terrible noisy, wrong, and late sensors don’t measure what you want

Think: Computation and AI

Think build a model of the world from your sensor data figure out what to do to get you to the state of the world you want implemented in software everything boils down to specific numbers

Interpreting Sensor Data

you have to write code input: (r, g, b) values for each pixel output: something you care about

A Simple Example turn on a light when someone is 1m from a LIDAR sensor

Another Example: Face Detection

Face Detection Viola-Jones algorithm Haar cascade

Set of Pixel Values that Often Correlate Well with the Presence of Faces in the Training Data that You Collected Detector Works great when it works Fails strangely when it fails

Figuring Out What To Do map every state of the world to an action in software

A Simple Example stop the robot 1m from the wall LIDAR sensor

Sensor Filters stabilize the readings introduce lag and bias

Implementing it in Software you have to pin down everything specific numbers if then you can’t test it all

Act: Actuator Technology

Moving About

Picking things Up

What Else is Hard?

Interacting with humans subtle and quick to anger Power everything boils down to battery life A bunch of other technical stuff

How to Watch a Robot Video

Folding Towels

Crazy Quadcopters

Final Thoughts a long way from the lab to the world easy for us usually means hard for robots the best way to understand modern robots is to use one and try to program it I understand law and policy less than you understand robots (so we should talk, over a beer)