Google Lunar XPRIZE – 4-kg Lunar Rover

One of the coolest classes at CMU is the 16-861 Mobile Robot Development/16-865 Advanced Mobile Robot Development (of which I am now the TA). Even though CMU had formally dropped out of the competition, when I took the class, the goal was still to design a rover for the Google Lunar XPRIZE.


One of the goals for the Google Lunar XPRIZE was to design a rover under 4-kg. Our ride to the moon, Astrobotic, Inc.’s Lunar Lander, specified that any payloads on their lander needed to be within a 40x33x38-cm envelope. This means designing things as simple and compact as possible. Because of my affinity for mechanisms, I decided to tackle the suspension/body averaging system. The previous semester of the class came up with a drivetrain design where the driven gears have the potential to also act as wheels and adding a gear reduction requiring a smaller motor, potentially saving on mass, shown below:


This design also allows for mobility using only two motors as part of a skid-steer system. A suitable suspension system for this design is a body averaging system, or “rocker bogie” system that is similar to what Curiosity uses. Applied to this robot, as one side tilts up, the other side of the robot tilts down, “averaging” the position of the body.

Rocker-bogie system on Curiosity. Source:

Unlike on Curiosity which has individually driven wheels, this rover only has two motors, driving two wheels each. This created an interesting design problem, how do you design a suspension where the motor is fixed, but allows for rotation around the same axis of the shaft of the motor?


This was a puzzle that took a while to solve, but a fun puzzle nonetheless. The body averaging system worked like this:


Cross-section of mechanism shown above.
Screen Shot 2018-02-17 at 11.35.01 PM.png
Labeled cross section of mechanism. Slightly different than what is shown above because it integrated the final CAD from other members of the team.

Build and Test

Machined prototype mechanism
Prototyped mechanism with 3D printed mount and laser-cut rocker arm
CNC’d rocker arm for the final prototype. I developed the G-Code in Mastercam and set-up the CNC (with some help from the machinists) myself.


Only broke one bit, and on the last pass, so close!
It came alive!


This class taught me a little more about designing for space and allowed me to design and machine (and use the CNC for the first time) to build my mechanism. It was definitely cool to see the whole thing turning with the gears that someone else in the class designed and machined. It would have been very nice to continue and refine this, but another project came along, this time shooting for a 2-kg lunar rover. While this mechanism didn’t make it into the new design, I was able to use my knowledge about the entire system as the lead systems engineer, which you can read about here.

The mechanical squad at the final presentation, ft. my crazy mountain man beard
Assembled final prototype in the Mars dirt simulant testbed at CMU.
2017.10.14 ZZZZDSC07827.JPG.jpg
Generations of CMU’s rovers, from Andy, down to the midterm 2-kg design described here.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s