Note: These images are very rough and not entirely to scale.
They are just to provide the basic idea.
Here is a list and perlimanery details about any upcoming projects the I have. If you have experiance with any projects similar to these, advice and suggestions, or you know where I can get some guidance, please don't hesitate to contact me. Remember, any and assistance will be most welcome.
Select a Project:
Current Project - Plotter
Future Project - Insect
The current project is the project that is on the front burner at the moment, and the Future project is on the back burner. Note: When I say a "Distant Future Project" I mean a project likely to have a start date in the dim distant distant future, a long long time from now! I am in my final year at High School after all, and there's only 24 hours in a day not 26 like some hope.
This project might seem a little bit in the wrong place, see as this site is on robotics, but it is not. It is controlled by a computer, and while not mobile, responds to external simuli. It has sensors, and must track an object as it move. Besides once completed it will aid in the production of future electronics/robotics.
Note: These images are very rough and not entirely to scale.
They are just to provide the basic idea.
More Coming Soon!
Notes: This plotter is designed to be able to print on the largest size PCB board readily available in Australia. It is 30x30cm in size and is about $18 in cost from Jaycar Electronics ($20 in Dick Smith Electronics and Altronics). I know I will not need a PCB that size for a robot, but you never know. I might need to produce several smaller PCB at the one time and as I was once told: "You can't draw a 5m striaght line accuratly if your ruler is only 30cm long."
Notes: I know you could probably buy something like this, but it's likely to be quite pricey and I'm working on a shortstring (i.e. $0) budget. Besides I want to have the fun of blowing it up (relativly speaking) during testing or on it's first trial. Also it is as good as place as any to design, construct, and test, computer related electronics, which will come in handy in the future. As well, take into consideration that other members of my family are good at electronics and build things occasionally.
Back to Top
This project will take in many areas of robotics at the one time. I will be quite fun learning the small things as I go along.
There are two separte identities or units in this project. First there is the central computer, which will be fixed at my desk. It is likely to be a 486 and is the "Boss" unit. Then there is the actual robots themselves. These are known as the "Sub" units. If time and funds are not lacking, (however they are lacking at the moment), there will be many Sub units. They will comunicate to the Boss unit by micro power radio in a similar fashion to computer networks. The Sub unit are going to be able to identify each other, and report back to the Boss unit which other units it has identified and where that are relative to itself. Also if there is enough room, the Sub units will have a crude radar type device, enabling them to `see' their surroundings. This is also likly to be reported in a compressed form to the Boss unit.
I know this is ambitious and will likely have to scale back my designs a bit. I already have!
Ideas relating to this project so far:
Use ultrasound for identification - i.e. each Sub unit has a unique `code' transposed on the ultrasonic waveform. Advantages - Distance easy to calculate. Disadvantages - Circuit sensitive to externals, directional, short range, echos.
Use a Infra Red led flashing at a high speed and a `code' for identification instead. A I-R LDR could be used for detection. Advantages - Larger range, easier to decode. Disadvantages - Unable to tell distance. (I've heard that this can be found by comparing waveforms, but I have not a clue as to how it is done)
Communication system uses a radio transmitter in each Sub unit, all tuned to the one frequency, and two radio receivers. One is the `Send' channel from the Sub units to Boss unit, and one is the `Recevie' channel from Boss unit to the Sub units. This way the Boss unit can send a message saying two Sub units are tring to talk to it at once, so all units are to Shut up. Also the Boss unit could send a message saying that it is listening to a certain Sub for the next x amount of time. The Sub units would listen for `silence' on the `Send' channel and would then send a tone/s saying that it has something to say. The Boss unit would then respond by sending a tone saying it is now listening to that Sub unit. The Sub unit would then send a little bit of data to the Boss unit and then wait for another free peroid. The data that the Sub unit wants to send would be split up into little parcels with a checksum and these parcels would be what is sent. With this system I could have many Sub units and only need two frequencys, one in and one out. The plain disadvantage to the above scheme is that timing is critical, and the Boss unit would likly drown out any radio signals. But how else can you talk through several layers of wood?
More details and rough pictures soon.
Back to Top
That's all the project I have at the moment. If any new ones pop up, they will appear on this page quickly. After all, with the millions of people on the internet, surely someone out there will be able to offer some advice. As always, feel free to contact me if you spot an error, or you have something to say. I can assure you that all corespondence will be read and taken into consideration, but ... it might take some time before I respond to it. I will respond, even if it is just a thank you note, but I will in the end. Thanks in advance for understanding. Currently I'm contactable here:
Last Upated: 29/4/99