A self balancing robot - Autodesk

1 downloads 298 Views 3MB Size Report
Inventor in Motion project on github. Contact me! ▫ Mail: [email protected]. ▫ Twitter: @davidtruyens. ▫ G
How to Design, Simulate, and Manufacture a Self-Balancing Robot David Truyens Application Engineer Twitter: @davidtruyens

Join the conversation #AU2015

Key learning objectives At the end of this class, you will be able to:  Learn how to simulate a self-balancing robot within Inventor DS  Learn how to use this technology to solve real-world challenges  Discover what the speaker has tried and what didn't work  Discover the future of making machines

Why?

Why you shouldn’t …  Uncountable hours

 Money  Less sleep

 Headache  Relationship

Why you should!

 Controllers are everywhere!  Everything is possible these days  FOMT

Full Motion Dynamics

Flanders Make

Balanduino / TKJ Electronics

Arduino - Massimo Banzi

Gadgets

Hibbot

Ampelmann

Real stuff

Space and automotive

Why you should

A self balancing robot Control Software

Firmware

CAD

Hardware

Virtual

Physical

Theory

A self balancing robot

Firmware

Virtual

Physical

Garage version

Firmware

Virtual

Physical

Garage Version  Open Pandora’s box:

Garage Version

9 DoF MPU?  Gyro (electronical) = angular velocity Very accurate!

No angle!!

 Accelerometer = gravity and other accelerations Direct angle

Lot’s of noise…

 Compass = North direction in an XYZ vector Gravity independent

Slow and not accurate

PID Controller

I*

 P

proportional

 I

integral

 D

derivative

𝐷 ∗ 𝜃𝑏

𝑏𝑎𝑙𝑎𝑛𝑐𝑒 𝑒𝑟𝑟𝑜𝑟

𝑷 ∗ 𝜃𝑏

Master Slave - PID Controller 𝜃𝑇𝑎𝑟𝑔𝑒𝑡 = 𝑃𝑝𝑜𝑠 ∗ ∆𝑑 − 𝐷𝑝𝑜𝑠 ∗ 𝑣

Master-slave controller

0

+ -

Cpos 𝑑 𝑑

+ -

Cangle 𝜃 𝜃

Kalman

Marvin

Next challenge…. Gearboxes…

Master-slave fuzzy-logic controller

0

+ -

Cpos 𝑑 𝑑

+ -

Cangle

Marvin

𝜃 𝜃

Kalman

𝑃𝑎𝑛𝑔𝑙𝑒 𝐷𝑎𝑛𝑔𝑙𝑒 𝑃𝑎𝑛𝑔𝑙𝑒 𝐷𝑎𝑛𝑔𝑙𝑒 𝑃𝑝𝑜𝑠 𝐷𝑝𝑜𝑠 𝐾𝑎𝑙𝑚𝑎𝑛

Marvin Launch Pad

Garage Version Advantages

Disadvantages

• Bottom up – step by step

• Trial and error, error, error

• Quick results

• Not possible for large scale

• Not much theoretical background needed • Fun / educational

projects • Limited in complexity

Classical approach Mathlab – Simulink - Simmechanics Firmware

Virtual

Physical

Classical approach Advantages

Disadvantages

• Industry standard

• Broken workflow in Inventor

• Lots of options

• Complicated

• Option to link cad data

• Easy to make mistakes • No validation

Co-Simulation approach Mathlab – Cosimate - Inventor Firmware

Virtual

Physical

Co-Simulation approach

Co-Simulation approach Advantages • Connected worfklows with Inventor and Mathlab

Disadvantages • Slow • Complicated

• Industry standard

• Easy to make mistakes

• Collaborate with multiple

• No validation

users (Cosimate)

Virtual validation workflow

Firmware

Virtual

Physical

Virtual validation workflow

Virtual validation workflow Advantages

Disadvantages

• Fully integrated in Inventor

• Quite hidden

• Validation of the control

• Not easy to program

software • Lots of potential using iLogic

• Strange things when programming

Conclusions  Lots of way’s to work

 Make your hands dirty

Future  Collaborate with Autodesk on Smart Machines  Create my own pcb with Circuits IO

 Make the next Marvin

Next generation

Open call Inventor in Motion project on github Contact me!    

Mail: [email protected] Twitter: @davidtruyens Github: https://github.com/DavidTruyens Fusion model: http://a360.co/1XD0Xt4

Autodesk is a registered trademark of Autodesk, Inc., and/or its subsidiaries and/or affiliates in the USA and/or other countries. All other brand names, product names, or trademarks belong to their respective holders. Autodesk reserves the right to alter product and services offerings, and specifications and pricing at any time without notice, and is not responsible for typographical or graphical errors that may appear in this document. © 2015 Autodesk, Inc. All rights reserved.