markov crawler
play

Markov Crawler Machine learning platform aiding CS188 Crawler can - PowerPoint PPT Presentation

Jan 23, 2024 •463 likes •562 views

Markov Crawler Machine learning platform aiding CS188 Crawler can execute a reflex agent or Q-learn optimal policy using Markov decision process. Raspberry Pi runs CS188 AI Python software controlling crawler 2 servo motors for arm and hand,

  1. Markov Crawler Machine learning platform aiding CS188 Crawler can execute a reflex agent or Q-learn optimal policy using Markov decision process. Raspberry Pi runs CS188 AI Python software controlling crawler 2 servo motors for arm and hand, acquiring crawler position from optical mouse. All parts from eBay or 3D printed, assembly in few hours. Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 1

  2. Markov Crawler features 1. Crawler is controlled by an on-board Raspberry Pi running Rasbian, remote ssh is available via WiFi dongle, 2. It is integrated with Python code used in CS188 class, 3. The dimensions of the virtual model (top fig.) are adjusted to physical crawler (bot fig.), computed position matches to the actual one, 4. Crawler can execute reflex-type policy defined by the operator, 5. Crawler can Q-learn policy rewarded by movement of the body measured by the optical mouse 6. Crawler can follows Q-policy inferred from the real body position or from the virtual model Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 2

  3. List of key prefabricated parts ● Raspberry Pi SoC ● muSD memory card ● WiFi dongle ● mini optical mouse ● PWM generator ● standard 5V servo motor x2 ● DC-DC 5V/4A step-down voltage converter x2 ● usb plug to power R-Pi ● servos On/Off switch ● Ampere meter (optional) ● 120V AC-12V/4A DC power supply (not shown) ● flexible 3m long 12V power supply cable Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 3

  4. Electrical harness to be assembled Electrical assembly takes 2 hours for a person with moderate soldering experience. A single 12V/4A power supply is connected via flexible 2- wire conductor (not shown). Average power consumption 0.5A * 12V=6W ● R-Pi w/ WiFi & opt-mouse 0.7A * 5V=3.5W ● 2 servos peak total power 2A * 5V = 10W Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 4

  5. List of 3D printed parts Upper Body Undercarriage Arm and Hand Parts Parts J H J C E G D F V U A B Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 5

  6. Mechanical assembly Mechanical assembly takes 2 hours, requires glue-gun, screwdriver, utility knife. Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 6

  7. Markov Crawler top view Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 7

  8. Connection to R-Pi via WiFi or directly R-Pi works as regular a PC with keyboard, mouse, screen, and copper Ethernet plugged in. A temporary direct connection to R-Pi (shown here) helps in the low-level debugging. Markov Crawler • jan.balewski@gmail.com • (812)-606-5607 • www.linkedin.com/in/janbalewski/ 8

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

SCORPION B SCAN CRAWLER What is the Scorpion B Scan Crawler? The Scorpion is a rugged

SCORPION B SCAN CRAWLER What is the Scorpion B Scan Crawler? The Scorpion is a rugged

SCORPION B SCAN CRAWLER What is the Scorpion B Scan Crawler? The Scorpion is a rugged remote access ultrasonic crawler designed to allow cost Effective A and B-scan imaging on above ground ferro-magnetic structures without the need for

387 views • 3 slides
Slug: A Semantic Web Crawler Leigh Dodds Engineering Manager, Ingenta Jena User Conference May

Slug: A Semantic Web Crawler Leigh Dodds Engineering Manager, Ingenta Jena User Conference May

Slug: A Semantic Web Crawler Leigh Dodds Engineering Manager, Ingenta Jena User Conference May 2006 Overview Do we need Semantic Web Crawlers? Current Features Crawler Architecture Crawler Configuration Applications and

502 views • 27 slides
Markov processes (Markov chains) Construct a Bayes net from these variables: parents? Markov

Markov processes (Markov chains) Construct a Bayes net from these variables: parents? Markov

Markov processes (Markov chains) Construct a Bayes net from these variables: parents? Markov assumption: X t depends on bounded subset of X 0: t 1 Temporal probability models First-order Markov process: P ( X t | X 0: t 1 ) = P ( X t | X t

359 views • 7 slides
UbiCrawler: a scalable fully distributed web crawler Paolo Boldi, Bruno Codenotti, Massimo

UbiCrawler: a scalable fully distributed web crawler Paolo Boldi, Bruno Codenotti, Massimo

UbiCrawler: a scalable fully distributed web crawler Paolo Boldi, Bruno Codenotti, Massimo Santini and Sebastiano Vigna 27th January 2003 Abstract We report our experience in implementing UbiCrawler, a scalable distributed web crawler, using

211 views • 18 slides
Markov chains and Hidden Markov Models 9000 Markov chains and HMMs We will discuss: Markov

Markov chains and Hidden Markov Models 9000 Markov chains and HMMs We will discuss: Markov

Markov chains and Hidden Markov Models 9000 Markov chains and HMMs We will discuss: Markov chains Hidden Markov Models (HMMs) Algorithms: Viterbi, forward, backward, posterior decoding Profile HMMs Baum-Welch algorithm 9001

1.16k views • 87 slides
Markov Systems, Markov Decision Processes, and Dynamic Programming Andrew W. Moore Note to

Markov Systems, Markov Decision Processes, and Dynamic Programming Andrew W. Moore Note to

Artificial Intelligence 15-381 April 5, 2007 Sequential Decision Problems & Markov Decision Processes Recap of last lecture Reasoning over time - Markov Processes - Hidden Markov Models - modeling state transitions - probability of

321 views • 16 slides
Markov Chains Markov Processes Discrete-time Markov Chains Continuous-time Markov Chains Dr

Markov Chains Markov Processes Discrete-time Markov Chains Continuous-time Markov Chains Dr

Markov Chains Markov Processes Discrete-time Markov Chains Continuous-time Markov Chains Dr Conor McArdle EE414 - Markov Chains 1/30 Markov Processes A Markov Process is a stochastic process X t with the Markov property : Pr ( X t n x n |

491 views • 30 slides
1 A Crawler Architecture Web Crawler Starts with a set of seeds Seeds are added to a URL

1 A Crawler Architecture Web Crawler Starts with a set of seeds Seeds are added to a URL

Table of Content Basic crawling architecture and flow Distributed crawling Scheduling: Where to crawl Crawling Crawling control with robots.txt Freshness Focused crawling URL discovery T. Yang, UCSB 290N Deep web,

379 views • 8 slides
Markov Logic Markov Logic Probability First-Order Logic Propositional Logic Markov Logic

Markov Logic Markov Logic Probability First-Order Logic Propositional Logic Markov Logic

Putting the Pieces Together Markov Logic Markov Logic Probability First-Order Logic Propositional Logic Markov Logic Definition A Markov Logic Network (MLN) is a set of A logical KB is a set of hard constraints pairs (F, w) where on

350 views • 6 slides
Outline Markov networks (a.k.a. Markov random fields) Markov Networks Reading: Michael

Outline Markov networks (a.k.a. Markov random fields) Markov Networks Reading: Michael

Outline Markov networks (a.k.a. Markov random fields) Markov Networks Reading: Michael Jordan, Graphical Models , Statistical Science (Special November 12, 2009 Issue on Bayesian Statistics), 19, 140- CS 486/686 155, 2004. University

325 views • 4 slides
Crawling T. Yang, UCSB 293S Some of slides from Crofter/Metzler/Strohmans textbook Where are

Crawling T. Yang, UCSB 293S Some of slides from Crofter/Metzler/Strohmans textbook Where are

Crawling T. Yang, UCSB 293S Some of slides from Crofter/Metzler/Strohmans textbook Where are we? Internet Web documents Crawler Crawler Crawler Document Document Document respository respository Online respository Inverted index

748 views • 47 slides
1 X 1 X 2 X 3 Ghostbusters HMM Chain Rule and HMMs E 1 E 2 E 3 P(X 1 ) = uniform 1/9 1/9

1 X 1 X 2 X 3 Ghostbusters HMM Chain Rule and HMMs E 1 E 2 E 3 P(X 1 ) = uniform 1/9 1/9

Hidden Markov Models 1 Hidden Markov Models Markov chains not so useful for most agents Need observations to update your beliefs Hidden Markov models (HMMs) Underlying Markov chain over states X You observe outputs (effects) at

617 views • 7 slides
Crawler.NET: A component-based distributed framework for web traversal Levente Hunyadi (BME

Crawler.NET: A component-based distributed framework for web traversal Levente Hunyadi (BME

Crawler.NET: A component-based distributed framework for web traversal Levente Hunyadi (BME AAIT) March 23, 2007 Crawler.NET A component-based distributed framework for web traversal Motivation Introduction The Web: Motivation a source

601 views • 27 slides
CSE P 590 A Markov Models and Hidden Markov Models

CSE P 590 A Markov Models and Hidden Markov Models

CSE P 590 A Markov Models and Hidden Markov Models http://upload.wikimedia.org/wikipedia/commons/b/ba/Calico_cat Dosage Compensation Reminder: Proteins Read DNA and X-Inactivation E.g.: 2 copies (mom/dad) of each

1.28k views • 21 slides
Hidden Markov Models Pratik Lahiri Introduction A hidden Markov model (HMM) is a

Hidden Markov Models Pratik Lahiri Introduction A hidden Markov model (HMM) is a

Hidden Markov Models Pratik Lahiri Introduction A hidden Markov model (HMM) is a statistical Markov model in which the system being modeled is assumed to be a Markov process with unobserved (hidden) states. We call the observed event

741 views • 13 slides
Markov decision processes and interval Markov chains: exploiting the connection Mingmei Teo

Markov decision processes and interval Markov chains: exploiting the connection Mingmei Teo

Markov decision processes and interval Markov chains: exploiting the connection Mingmei Teo Supervisors: Prof. Nigel Bean, Dr Joshua Ross University of Adelaide July 10, 2013 Background Intervals Markov Decision Processes Markov chains

528 views • 30 slides
Thinking about thinking How children learn to talk about mental states Micha Elsner

Thinking about thinking How children learn to talk about mental states Micha Elsner

Thinking about thinking How children learn to talk about mental states Micha Elsner (Department of Linguistics) Why think about think? To learn: How linguists study infant development What we still dont know And

641 views • 39 slides
LCCC Workshop: System Design Meets Equation-based Languages Sep 20, 2012 Carl D. Laird, Assistant

LCCC Workshop: System Design Meets Equation-based Languages Sep 20, 2012 Carl D. Laird, Assistant

LCCC Workshop: System Design Meets Equation-based Languages Sep 20, 2012 Carl D. Laird, Assistant Professor Chemical Engineering, Texas A&M University William E. Hart, Jean-Paul Watson, John D. Siirola Sandia National Laboratories,

656 views • 47 slides
Enterprise Storage Architecture Fall 2018 Introduction Tyler Bletsch Duke University Slides

Enterprise Storage Architecture Fall 2018 Introduction Tyler Bletsch Duke University Slides

ECE590 Enterprise Storage Architecture Fall 2018 Introduction Tyler Bletsch Duke University Slides include material from Vince Freeh (NCSU) Instructor and TAs Professor: Tyler Bletsch Office: Hudson Hall 106 Email:

834 views • 30 slides
Model Answer for Assignment 2 In a factory, there are 3 component makers. Each maker produces a

Model Answer for Assignment 2 In a factory, there are 3 component makers. Each maker produces a

CMPT-300 02-01 E2 Model Answer for Assignment 2 In a factory, there are 3 component makers. Each maker produces a unique kind of components independently. All these components are stored in a moving belt with size 10 forwarding to the

516 views • 3 slides
Multiresolution Gaussian Processes Emily Fox ICERM 2012 Providence, RI

Multiresolution Gaussian Processes Emily Fox ICERM 2012 Providence, RI

Multiresolution Gaussian Processes Emily Fox ICERM 2012 Providence, RI Joint work with David Dunson (Duke) Goals Data from Neuronal Recordings Many :me series exhibit: 0.5 Long-range

1.2k views • 85 slides
Summary Individuals and interactions over processes and tools. This statement is one of the

Summary Individuals and interactions over processes and tools. This statement is one of the

How to use tools without losing focus on individuals and interactions Summary Individuals and interactions over processes and tools. This statement is one of the most important Agile values, and one I subscribe to. But tooling cannot be

741 views • 24 slides
Amplifier Lab Introduction: In this laboratory project you will build an amplifier circuit. This

Amplifier Lab Introduction: In this laboratory project you will build an amplifier circuit. This

K. Disney Engr 003 Mission College Amplifier Lab Introduction: In this laboratory project you will build an amplifier circuit. This amplifier is able to increase the strength of a sound signal. The amplifier can be used to amplify the weak

483 views • 13 slides
October 19, 2009 Current Issues Wi-Fi Tracking Systems Infrared Cameras Required

October 19, 2009 Current Issues Wi-Fi Tracking Systems Infrared Cameras Required

October 19, 2009 Current Issues Wi-Fi Tracking Systems Infrared Cameras Required Sensors Wii Remote Payload Testing Status Tasks for this week Wii Remote: MicroCART Weight close to max payload

360 views • 11 slides

More recommend