Go Down

Topic: Step-up Projects to Tracked Gas Detector (Read 1 time) previous topic - next topic


Ladies and Gentlemen,

Some background:
I am currently a newbie, and the Popular Mechanics "Build Your First Robot" caught my eye.
I am a Volunteer Firefighter.
I almost majored in Computer Science, and am very familiar with C++.

The bottom-line up-front: I am looking for solid, well-developed, well-documented, intermidiate projects that will help me develop the skills and knowledge to create the following:

I am looking to make a robot that is capable of moving through a structure and detect levels of cabon monoxide and other gases.  The end result would be a robot that (if allowed by departmental policy, which it's not - this is purely a personal project) could move into a potentially IDLH (immediately dangerous to life or health) environment and not put firefighters at risk.  Initially, the robot does not need to be intrinsicly safe, but I would want to move there.

Looking at a tracked base that is long enough to tackle stairs, with tracks that are tall enough to allow the vehicle to flip on its "back" and adjust and continue as if right-side up.

Mandatory features:
CO detector

Optional Features/Capabilities:
Ability to send data wirelessly to a readout
Audible warning when IDLH atmosphere detected
Multiple other gas meters (methane, hydrogen, etc.)
O2 meter
Ability to "follow" a hoseline or tagline (rope)
Ability to send video over radio (I will have my HAM license well before I figure this out) to PC
Ability to receive commands over radio from PC (I want the robot to be able to operate autonomously and "explore", but also be able to be directed to move to/observe other areas)
LED "bar graph" and speaker to indicate levels of gases which require the attention of nearby personnel

I am aware of how ambitious this project is at my current skill level.  Accordingly, I am planning on conducting several step-up projects to reach this point.  

The "Build Your First Robot" Project: this will develop my building capabilities, introduce me to the programming system, learn wheeled locomotion, collision avoidance (switch-based)

A Gas Meter project: the project I've seen will convert the readout and display on the computer monitor

A display project: taking the information from the sensor and placing it on a Arduino readout

Projects I know I need to explore:

Interfacing the Arduino with HAM radio
Tracked operation
Non-contact collision avoidance
Interfacing R/C and Arduino (I believe the coding should be fairly simple if (command exists) then (follow it) else (explore autonomously)

Whatever help and direction the community can give me would be greatly appreciated.


Yeah, sounds like a project that commercial robotics companies, like iRobot, have
spent millions of $$ and multiple manyears developing.


Building a robot that can do something as seemingly mundane as climbing stairs has been
a major challenge right there. Also, coming up with nice control interface GUIs is another
major task.

I'll recommend one thing, the following guys make a whole series of gas detectors.


The best advice is get some off the shelf robotics platform and start there to learn the
ropes. The most important thing is ... your first robot is not gonna be your last robot
[if you don't fizzle out on the 1st one], and most people build many of them, bootstrapping
learning curve off of each one to design the next one.


Mechanically climbing stairs is a problem.  Given the weird air currents in a house finding the source of a particular gas may be difficult also.  If you got the sniffer down maybe you could throw the robots through a window on each floor and they could search out the gas source out on that level.  I worked on small UAVs and know that most, probably all, electronics used in a robot can easily take the shock that going through a window would cause.


I am more conerned about IDLH. Are there any things done yet to prevent this?

Always assume hazardous atmosphere (it can change at any time) and go in on air. As to the project, an interesting personal exercise, but if it were practical, the bots would already exist. Below might be an inexpensive place to start experimenting with the project.

Google forum search: Use Google Search box in upper right side of this page.
Why I like my 2005 Rio Yellow Honda S2000  https://www.youtube.com/watch?v=pWjMvrkUqX0


I appreciate the input thus far.  I know I haven't posted in my own thread, but I have been reading the posts here, as well as continuing to delve into this concept.

I'm starting with a little tracked Tamiya chasis, making a proof of concept working while I'm here (deployed to Afghanistan) that I can scale up once I get home.  Still waiting on some parts, but I've been able to play around.

A question that I've come up against is how, if I get this to work, to make it able to go through decon.  I haven't played around a lot with electronics before, so I don't know if waterproof sensors exist, or if I could just power down the sensor board (final project will be two independent Arduinos, one running sensors and displays, the other navigation) while going through decon, and then just wait for it to dry.  Everything else could be easily waterproofed.

Also, does an actual O2 sensor compatible with the Arduino exist?  I don't need a dissolved oxygen sensor, I'm looking for one that says "current air O2 is 15%" so the board can say "15% is IDLH".  So far my searches have been fruitless, I didn't know if anyone knew where I could find one.

Go Up