Mars Rover Simulator and Augmented Model

From Makers Local 256
Revision as of 22:48, 5 November 2012 by Tylercrumpton (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Brainstorming (paused)
Born On:
18:08, 14 August 2012 (CDT)
Last Updated:
22:48, 05 November 2012 (CDT)


The Mars Rover Simulator and Augmented Model, or "MR.SaAM" for short, is a contained, controllable system that allows a user to pilot a tiny rover robot in a small Mars-like environment.


Currently, the following people are working on this project:


Each task is a separate, and mostly independent, component of the project as a whole. There is no particular order that the tasks need to be completed in.


This task involves building a small, contained model-environment that is composed of Mars-like land features. We will probably use one of the coffee tables at the shop, while ensuring that the table still functions as a coffee table.

  • Colored sand and rocks for the terrain
  • Hidden signs of "life"
  • Crazy-awesome aperture opening for robot entrance and exit
  • Maybe some sort of lighting (or this could be handled by a light on the rover)
  • Use a bumper-car style mesh system to power the rover (might be hard to make a connection through the sand to the grounding floor)


This is the rover itself. We need to decide on a style of rover (Curiosity, Spirit, Opportunity, custom design?), features (manipulators, lights, camera), and mechatronics (power system, control system, mechanical).

  • Some sort of grasper or manipulator to move rocks and stuff around
  • A flashlight if no environmental lighting is used
  • Laser-cut/3D-printed chassis, so we can open-source the design?
  • Power options:
    • Battery-powered (would require recharging, take up valuable rover space)
    • Bumper-car Mesh (use the ceiling and/or floor to provide power to the rover)
    • Tethered (not ideal, just have a long wire power the rover; it would help with data and video comm, though)


The rover will be controlled by some sort of control application.

  • Programmable communications delay, based on distance from Earth to Mars
  • Fancy NASA control interface
  • Web-enabled, so anyone can control the robot from a computer or mobile device
  • Option to remove/edit the delay, if you just want to drive around
  • Robot and environment status alerts (low-battery, sensor data)


  • Had a conversation with Tyler yesterday and I think we determined that working on the robot design first would be the most feasible. This would give us an idea of the size of robot possible and scale other things accordingly. Building it around on the the wireless enabled microcontrollers seemed like the best plan. It would communicate with the middleman computer on or near the table and the middleman computer would talk to the net. Tyler showed me some rather small servos that might work fairly well if they were modified for continuous rotation. Should be able to calculate distance travel per rotation so the software commands can be queued similar to how current rovers are done. --strages (talk) 10:18, 17 August 2012 (CDT)
    • I have a cheap little thin-client computer that would probably work well for the middle-man computer that I've never really had a good use for. Also, the servos can most certainly be modified for continuous rotation, and it's not a bad process at all to do so. I've modded one, and have the correct resistors to make plenty more. Unfortunately, you lose positional feedback when you modify them, so we would lose that functionality. --Tylercrumpton (talk) 21:58, 19 August 2012 (CDT)
  • See the Ardurover project. This may be more than what you want to accomplish, but an off-the-shelf solution frees up resources for more ambitious goals. --krs (talk) 14:18, 17 August 2012 (CDT)
    • I have actually messed with Ardurover a little (helped a friend get a project up and running), and while really awesome, I believe you're right in thinking that it might be a little overkill. I don't think this project will need too much autonomous navigation and telemetry, so I think we can get by with a simpler (and more importantly, cheaper) board/solution, without much extra work. Any thoughts, strages? --Tylercrumpton (talk) 22:04, 19 August 2012 (CDT)