Mecanum%20Omnidrive:%20What%20is%20it,%20how%20we%20did%20it%20and%20how%20well%20it%20worked

1
Mecanum Omnidrive What is it, how we did it and
how well it worked
Chad Hine Eastman Kodak Company MAK Robotics
Team 211 Rochester, New York 14650-2110, USA
2
Outline

• MAK Robotics Team
• This Years Game
• Why We Used Omnidrive?
• Why We Used Mecanum Omnidrive?
• What Is It?
• How Did We Build It?
• How Did We Control It?
• How Well Did It Work?
• Lessons Learned
• Questions

3
MAK Robotics Team

• Kodak first sponsored a team in 1994 with two
  other local high schools.
• In 1999, MAK (Marshall And Kodak) was formed.
• We hold fall seminars and tutorials to teach the
  students about robotics.
• We have won numerous awards for quality and
  innovative designs.
• We have between 10 and 15 volunteers from Kodak
  on the team.
• There are currently, 20-30 students from Marshall
  each year.
• We also have 5 teachers/administrators helping
  from Marshall.

4
This Years Game

• Crowded field with 6 robots and 9 goals.

5
Why We Used Omnidrive?

• We wanted something different! Not just the same
  old tetra stacker.
• Robot maneuverability was voted on as the top
  priority (over speed and power).
• Flat playing field allowed many different
  omnidrive designs.
• Kit finally included four matching motors capable
  enough for drive wheels.
• Looked like a fun challenge.
• We like winning awards.

6
Why We Used Mecanum Omnidrive?

• Brainstormed various ways of doing omnidrive.
• Ideas included regular omni wheels mounted at
  45 degree angles, tank turret, wheel turning,
  tripod-omnidrive, and more.
• Only major difference with Mecanum over standard
  drivetrain is the wheels.
• Mecanum was overall the quickest and easiest to
  do given enough manufacturing capability to be
  able to make the complicated wheels.
• Mecanum gave us the most options in case it did
  not work (just replace the wheels with standard
  wheels).
• Mecanum was also the lightest of the omnidrive
  solutions.
• They looked really cool (important to students).

7
What Is It?

• It is a way of driving a platform in any
  direction, without slowing or turning.
• It uses omniwheels with rollers mounted at a 45
  degree angle from the main wheel.
• Software controls how quickly each of the four
  wheels move in any direction.

8
What Is It?
Standard Omniwheel Mecanum Omniwheel
9
What Is It?
10
What Is It?
11
What Is It?
12
How Did We Build It?

• Used http//robotics.ee.uwa.edu.au/eyebot/doc/robo
  ts/omni.html

13
How Did We Build It?

• There is nothing special about the base other
  than the wheels being mounted equidistant from
  each other. This is not necessary, but greatly
  simplifies the software.
• The wheels were completely designed in UG by a
  team member.
• The hubs were created by stereolithography using
  nylon.
• The shafts were manufactured by another team
  member.
• The cone rollers were molded in the mold shop
  using polyurethane.
• The center rollers were purchased.
• The wheels were completely assembled by the
  students.

14
How Did We Build It?
15
How Did We Control It?

• Used a single joystick.
• When you push forward, it went forward. When you
  push left, it went left. When you push diagonal,
  it went diagonal. etc.
• 8 different wheel direction modes

• Slope of Yellow dot says we are going diagonal.
• Since X and Y are positive, we know we are going
  forward-right diagonal.
• Distance from origin is how fast we will travel
  in that direction.
• Red square is dead zone meaning dont move.

16
How Well Did It Work?
Realistically, six weeks to design, build and
test something this new and innovative is not
nearly enough time.

• We had major problems controlling the robot.
• We had robustness issues with the stereolithed
  hubs.
• We had very bad power and speed in the
  cork-screw directions.
• We had minor problems with the small wheel base
  this design forced on us.
• Overall, our robot placed very well.
• We had very good power and speed in the
  straight directions.
• Our drivetrain won us Most Innovative Design
  Award.
• Our uniqueness earned us spots in the playoffs in
  both competitions we attended.

17
How Well Did It Work?
18
Lessons Learned

• Try to make all 4 drive motors face the same way
  to negate motor bias.
• Spend much more time on the software algorithms.
• Use LUTs to help keep motors around same speed.
• Use sensing devices and feedback to force the
  wheels to drive at the exact same speed.
• Make the stereolithe hubs more robust.
• Adding an individual suspension system to each
  wheel would greatly help over rough surfaces.

19
Lessons Learned
20
Questions?
21
Thank You

• To all the hard working Kodak volunteers that
  made this possible
• John Pochan Charles Patton Jaime Lange
• Norm Morrill Jim Olds Scott Phillips
• Joe Kabasta Don Hammen Mike Schafer
• Joanne Begy Jeff Lange
• To our generous sponsors who make this possible
  every year
• Glen Pearson
• George Wacker
• Jim Stoffel
• Essie Calhoun

22
Thank You