Introduction to Robotics Shivam Goel A bit about me.. PhD student - PowerPoint PPT Presentation

Introduction to Robotics Shivam Goel

A bit about me.. • PhD student • Advised by Dr. Diane J. Cook and Dr. Matthew E. Taylor • Research interests: Computer vision, Reinforcement learning, Robotics

Outline • Introduction • History of Robotics • Successes and failure • Robot mapping • Simultaneous Localization and mapping (SLAM) • Some cool applications • Summary and Future directions

History of robotics • In 1495, Leonardo da Vinci drew plans for a mechanical man. The model of Leonardo Da Vinci's robot with inner workings as displayed in Berlin

History of Robotics • The acclaimed Czech playwright Karel Capek (1890-1938) made the first use of the word ‘robot’, from the Czech word for forced labor or serf.

History of Robotics • Science Fiction writer who moved into Robotics. Was the first used the term "robotics' to refer to the study of robotic applications.

Definition of Robotics • A reprogrammable, multifunctional manipulator designed to move material, parts, tools, or specialized devices through various programmed motions for the performance of a variety of tasks.

Robotics: Successes • Space exploration: Mars Rover, Spirit • Self-driving cars: Google, Waymo, Uber • Household robotics: Roomba • Industrial robotics: KUKA

Robotics: Failures • Robots in Space: Robonaut 2 • Self-driving cars: Uber Self driving car kills a pedestrian in Tempe, AZ • Some funny robot fails

Outline • Introduction • History of Robotics • Successes and failures • Robot mapping • Simultaneous Localization and mapping (SLAM) • Some cool applications • Summary and Future directions

Robot mapping Real World Environment Motion Control Perception Environment Model, Path Local Map Position Global Map Cognition Localization

Robot mapping Perception Computer Vision Haptics Natural Language Processing

Robot mapping Motion Control Environment Model, Path Local Map Position Global Map Cognition Localization

Localization example

Localization • Two types of approaches: • Iconic : use raw sensor data directly. Match current sensor readings with what was observed in the past • Feature-based : extract features of the environment, such as corners and doorways. Match current observations

Mapping example

SLAM example

SLAM: applications Indoors Undersea Underground Space

SLAM • Simultaneous localization and mapping: Is it possible for a mobile robot to be placed at an unknown location in an unknown environment and for the robot to incrementally build a consistent map of this environment while simultaneously determining its location within this map? http://flic.kr/p/9jdHrL

The SLAM problem • SLAM is a chicken-or-egg problem: • A map is needed for localizing the robot. • A pose estimate is needed to build the map. • Thus, SLAM is regarded as a hard problem in robotics

Three Basic Steps • The robot moves • increases the uncertainty on robot pose • need a mathematical model for the motion • called motion model

Three Basic Steps • The robot discovers interesting features in the environment • called landmarks • uncertainty in the location of landmarks • need a mathematical model to determine the position of the landmarks from sensor data • called inverse observation model

Three Basic Steps • The robot observes previously mapped landmarks • uses them to correct both self localization and the localization of all landmarks in space • uncertainties decrease • need a model to predict the measurement from predicted landmark location and robot localization • called direct observation model

How to do SLAM

How to do SLAM

How to do SLAM

How to do SLAM

How to do SLAM

How to do SLAM

How to do SLAM

How to do SLAM

How to do SLAM

*Slides adapted from Cyrill Stachniss

*Slides adapted from Cyrill Stachniss

Three main SLAM paradigms • Kalman filter • Particle filter • Graph based

Graphical Model of Full SLAM: p ( x , m | z , u ) Arrows = influences 1 : t 1 : t 1 : t *Slides adapted from Cyrill Stachniss

Graphical Model of Online SLAM: òò ò = p ( x , m | z , u ) ! p ( x , m | z , u ) dx dx ... dx - t 1 : t 1 : t 1 : t 1 : t 1 : t 1 2 t 1 *Slides adapted from Cyrill Stachniss 39

SLAM: Demo • Video

Outline • Introduction • History of Robotics • Successes and failures • Robot mapping • Simultaneous Localization and mapping (SLAM) • Some cool applications • Summary and Future directions

Bin Dog: Intelligent In-orchard Bin-Managing System for Tree fruit production Center for Robotic Decision Precision & Automated Making Laboratory Agricultural Systems

Overview of Bin-Dog Hardware Basic specifications of bin dog system Track 1.55 m Wheelbase 1.95 m Height (collapsed) 1.5 m Height (Fully extended) 2.1 m Engine Power 9.6 kW Max Speed 1.2 m·s -1 Max Steering rate 30 deg·s -1 Capacity of bin-loading 500 kg system o Four-wheel-independent steering system (4WIS) o Passive suspension o Bin-loading system 43

Bin-Dog Autonomy Block diagram of bin-dog navigation system Data processing in Robot Operating System

Bin-Dog Autonomy q GPS-based navigation system o Localization and path planning Sensors (GPS & IMU) q Actual trajectories Enter an alleyway

Bin-Dog Autonomy q Laser-based navigation system Detection of alleyway entrance

Bin-Dog Autonomy q Laser-based navigation system Detection of tree rows Detection of bin in an alleyway

Bin-Dog Autonomy q Laser-based navigation system Detection of tree rows Position Error computation

Bin-Dog: showcase Fetch a target bin

Bin-Dog: showcase q Automated bin management Place an empty bin

Bin-Dog: showcase Replace full bin with empty bin

Robot Activity Support (RAS) 52

Robot Activity Support (RAS) Camera module Tablet interface Navigation module Wilson, Garrett, et al. "Robot-enabled support of daily activities in smart home environments." Cognitive Systems Research 54 (2019): 258-272. 53

SLAM in RAS • Navigation module built using Google’s Cartographer system • SLAM system builds the map of the environment and determines the robot’s location on the map

RAS: showcase

RAS: showcase

Future of Robotics Source: https://www.therobotreport.com/10-biggest-challenges-in-robotics/

Summary • History of robotics dates to 1495 • In the past two decades robotics has seen many advances as well as some failures. • Robot mapping is a crucial aspect of mobile robotics • SLAM = simultaneous localization and mapping • Kalman filter, particle filter, graph based • Full SLAM vs Online SLAM • Resources to study more about SLAM • Online lecture video series by Cyrill Stachniss • Springer “Handbook of Robotics”, Chapter on SLAM • Introduction to Mobile Robotics, Chapters 6 and 7

Thank you