Let's Make Robots!

Mr. Radioactivity

Navigates around via Sharp and tactile sensor, measures ionizing radiation, telemetry of radioactive activity via wireless 2 digit counter
wireless_counter_schematic.JPG205.79 KB
bq-bot_program_1.bas660 bytes
7_segment_dispaly_counter_with_08M.JPG208.26 KB
Counter_program1.bas164 bytes

Hi everybody! Finally I found the time to start building Mr. Radioactivity. The robot navigates around via Sharp and tactile sensor, measures ionizing radiation via Geiger-Mueller tube and according circuit and sends the registered impules wireless to a 2 digit 7-segment counter. The RX/TX unit is from an old RC toy car. I removed the TX PCB from the transmitter case and placed the RX PCB as well as a small perfboard with two 4026 decade counter and 7-segment display driver in it (You can find the circuit diagram in the attachment):

The red transparent press button of the original transmitter case is now used as a cover for the 2 digit 7-segment displays:

Beside the wireless counter, the impulses will be ascertained by a Picaxe 08-M. If the nucleus decays per second exceed a defined level, the robot stops and emits an alarm sound.

I will use the 712 End Window-Alpha-Beta-Gamma Detector from Sparkfun as well as the circuit diagram mentioned here.

If the Sharp GP2D120 IR sensor detects an obstacle in an unacceptable range, the robot starts an evasive maneuver. The decision whether the robot turns left or right is random. If the number of registered nucleus decays, stored in a variable, at the moment of the obstacle detection is even, the robot turns left, if it is odd, the robot turns right.

I have written a short but effective test program for obstacle avoidance, using pseudo-random numbers (random command). Now I have to build the Geiger counter, unfortunately my Sparkfun order has following status since one week:

2010.05.26 My Sparkfun order has still the same status. I did some re-design and built a 4 wheeled (56 mm dia. TAMIYA) platform with 4 continues rotating servos which is more all terrain like:

2010.05.28 Some more progress on the new platform...

Tactile sensor, servo for IR and servo driver:

Battery case for MCU:

2010.05.29 Picaxe project board, H-bridge, voltage regulator board, Sharp IR mounted and test drive passed:

2010.05.29 High voltage generator for GM tube built and tested:

The circuit just draw 10 mA @ 5 V.

2010.06.10 I am still fighting with the Chinese customs to get my Geiger Mueller tube...Decided to mount the 2-digit 7 segment display counter on the robot. The two 7 segment displays show now directly the decays per second. Two CMOS 4026B control the 7 segment display. A Picaxe 08M evaluates the incoming pulses per second by the count command and clock/reset the CMOS 4026B accordingly. I have attached the circuit diagram and test code.

2010.06.14 Purchased a  cheap Chinese Geiger counter from taobao to check and calibrate my hombrew one. The counter is very sensitive (about 50 cpm) for β and γ. A test with KNO3 (isotopes of potassium are radioactive) shows an increase of around 3 bq.

2010.06.26 My father handed on a MR 9511 Geiger counter to me during my visit in Germany. As I opened the Geiger counter to change the battery I found that the MR 9511 has the same 712 End Window-Alpha-Beta-Gamma Detector which I recently ordered at Sparkfun and which is still in the Chinese customs. The MR 9511 has a 3.5 mm analoque output socket and the data sheet (in German) can be found here.

Tomorrow I will check some phosphate rocks and other uranium containing mineral.


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

WIth the nuclear deal in Japan have you got this out and measured.

as per our discussions...

I found two interesting notes/scripts regarding autonomous vacuum cleaner robots:

Wikipedia: Autonomous robots with random motion basically bounce off walls, whether those walls are sensed with physical bumpers like the Roomba cleaners or with electronic sensors like the Friendly Robotics lawn mower. The simple algorithm of bump and turn 30 degrees leads eventually to coverage of most or all of a floor or yard surface. [But without verfication]

Random Walk Application for Autonomous Vacuum Cleaner Robot

random enough

Schrodingers cat bot -

brilliant!.. ALL my robots are schrodinger's cat robots, they are simultaneously in a state of full functionality and miserable failure - until I turn them on and collapse the probabilty function.

oh - and where did you get the GM tube - and how do you supply the high voltage for it???

actually, have you mounted the GM tube yet? I don't think I can recognise it in the photographs.


but - why schrodinger's cat bot?

I will use the 712 End Window-Alpha-Beta-Gamma Detector from Sparkfun as well as the circuit diagram mentioned here. The order has costums problems, hanging since more than a week in Shanghai :(

The robot has similarity to the Schrödinger's cat thought experiment. The decision whether the robot turns left or right is random. If the number of registered nucleus decays, stored in a variable, at the moment of the obstacle detection is even, the robot turns left, if it is odd, the robot turns right. As we know, the decay of an unstable nucleus is entirely random and it is impossible to predict when a particular atom will decay...So it's undeterminable how the robot will react.

Furthermore the new platform remind me on a black small cat :)

well, you can probably ignore the following diatribe, it's just me being pedantic:

I think SC expt. had more to do with the cat ACTUALLY being in a mixture of states until the observation was made (i.e. both alive and dead - really) - rather than the system for generating the random number...? SC expt. works equally with any binary, true random generator, even a coin flip...besides, the GM tube doesn't actually measure the decay of anything, it only detects particles which may or may not have resulted from decay. The density of the particle shower is not random, and is entirely dependent on the environment. If you had a small source permanently attached to the GM tube, then that would be a better approximation of the randomness of decay events.

Another factor is that GM tubes are pretty lousy at detecting anything other than gamma radiation, which are pretty few and far between at sea level (depending on the voltage to your GM tube; a few detections every minute at most, unless you live at 3-mile island, or you have a small sample of isotope glued to your GM tube) but it certainly won't detect alpha particles which are too big and slow to get through more than a few cm of air - unless your source is RIGHT on the window.

attaching a small sample of isotope to the GM tube is not impossible,  you could hack a smoke detector to get the americum out. come to think of it, you could probably save yourself $100 and a lot of hassle by just hacking a smoke detector and pulling out the entire ionizing chamber + isotope in that.

Well, the robot is also in a mixture of states until the observation was made:) The mixture of states is called the Copenhagen interpretation :)

The Geiger tube has an end window and can detect α, β and γ. Normal background detection with this tube is about 30 clicks/minute. You will be surprised how radioactive our environment is! For example the dust, sticking on computer or TV screens.

The attamp of the robot is to search for radioactive sources, for instance pitchblende outdoors or the radioactive gas radon in basements. In the end a geiger counter on wheels:) To control the robots evasive maneuver with random numbers produced by nucleus decays, is only a side effect I am trying to use.

PS: A coin flip is not entirely random;-)

is it alive or no?

In my case the question is: turn the robot left or right?