Older blog entries for sashi_ono (starting at number 7)

7 Jun 2002 (updated 7 Jan 2003 at 05:35 UTC) »

I just got a hold of a Mark III robot kit. I plan on putting it together soon. The reason i got it was to get started programming on a pic 16f877. I need to pick up some pic assembly and I figured the best way is to put together a bot that uses the microcontroller that I want to learn. I'm thinking of using the 16f876 to use in my tinybot project. It should be small enough and use low enough power. The reason I haven't done it yet is because there is really no good (free) ide besides the microchip assembler. And I really hate writing code in assembly, I'd much rather use C. I mean I can always get a hold of a C compiler but then debugging and simulating won't be too fun because I haven't seen any good C debuggers that are free. So I'm back to learning pic assembly as convoluted as it is. I just really hate dealing with old backwards compatible technology, you know with all this paging crap. I'd wish they would start selling the 18f series soon. I just saw that they released it on their website but I haven't found a place to buy them. Well Anyways, hopefully the mark III kit goes together smoothly and I'll be on my way to pic assembly world.

I don't really like the pic all that much I'm thinking now of switching to an AVR. But i'm still not quite happy with that either. CAll me picky. I guess I am waiting for a good microcontroller to come out. I am placing on my wishlist a ARM7 microcontroller with flash, memory and all in a tiny package maybe 16 pins. i'd like a built in usb controller and serial I/O lines. blue tooth wouldn't be bad either. ALso a nice simple standard programming interface. JTAG? Anyone want to make one for me, or do i have to do it myself. Oh ya, and i want to pay less than 20$ a chip. Thanks to the first person to get it to me.

7 Sep 2001 (updated 7 Jan 2003 at 05:40 UTC) »

I think I've settled on a microcontroller for TinyBot. 16f876. It seems to provide everything I need in on package. it is also low power, has a/d , cheap :) and lots of previous code written for them. I will burn a bootloader onto it and use the serial port to do the code downloading and debugging. I just got a hold of an rs232 driver to do the downloading of the code. I also have access to a PIC programmer to download the bootloader. I downloaded the demo Hi_tech C compiler and got it compiling with MPLAB. seems like a good choice. hopefully I can keep my code small enough other wise the demo won't work. I could always break down and do it in asm but I'd rather not. Now the only thing left is to order them. looks like $8 from digikey. anyone know a good place to get them from aka cheap. let me know

I got hold of a Domino2 microcontroller. ANYONE knows anything about these things let me know. I got the spec sheets off the web and have been trying to connect to it via the serial port but no luck. If you havent seen one of these guys before, they are quite cool. they have a 80C52, 32k of sram, 32k eeprom 16I/O. built in regulator. And the best thing is that it is in a 1.5 cubic inch box. 1.5"x2"x.5". If anyone is interested they are made by micromint. The only thing is that they are costly like 150.

6 Jun 2001 (updated 4 Jan 2002 at 01:43 UTC) »

Words of wisdom from Sashi Ono.
How to build a robot

Step 1: Design
Outline your goals and write them down. Do not start anything until you have figured out what you want to accomplish. Then come up with your working design for your project. If you haven't done something similar before don't go too wild, keep it simple yet effective.

Step 2: Planning
Come up with a timeline of your project. Think carefully, as missing deadlines (contests) can mess up all your goals. Work in Parallel, that is do not waste time. If you are waiting for parts to come you should be doing something else. A simple breakup consists of Chassis, Hardware, and Software. I would recommend breaking up your progress into weekly steps to ensure constant progress. Make sure you get something big done once a week otherwise you will be crunched for time later on which is not good for your health.

Step 3: Ordering and Gathering Parts
This needs to be done as early as possible. Do not wait. Without parts your chassis and hardware will not progress. Do not waste your time here, pick something that works and go with it.

Step 4: Building
Hopefully you have gotten all your parts so you can start building and putting together your design. If you have planed well this step should be relatively simple just follow through with your design. My suggestion is to start with the Chassis as this determines the flow for the rest of the project.

Step 5: Debugging
This is where your engineering skills come out. You will run into problems. Expect it and plan for it. You may realize that your design doesn't work and may need to start over. The second time you do anything is faster. You will often need to come up with unique fixes and solutions.

Step 6: Optimizing
Once you get something working kind of the way you want you will probability want to optimize it. Feel free to do so but do not move far away from your initial goals and plans. An important skill is to keep you design constantly working. Do not give into the urge to dismantle your project for improving it. You are better off starting off new project and keeping your current design working as a fall back plan. Also know when to stop. Constant tweaking often creates problems and makes things messy and harder to debug when problems do happen.

Step 7: Show Off
After you have a working project feel free to show off. You have accomplished something big here and put a lot of time into it you have the right to. Encourage others to undertake what you have done if you feel that it was a worthwhile project. Do not bash others with similar projects as they have probability worked just as hard as you and you wouldn't want others to bash on you. Release documentation of your project to help others learn and gain from your project.

Ok figured out the best way to get the counter weights off. I crushed them with a vise. I put them in sideways and cranked it till the counter weights crumbled. I put together a motor solar panel assembley just to see how the solor panels handle a motor. Looks like i may be only able to power one motor at a time. I could put on some more panels. I'm also thinking of maybe putting on one AAA battery and using the solar cells to charge, that way i don't have to put more cells on. Still looking for a good cpu. Maybe a pic or something with builtin flash and ram and is somewhat fast.

I got my order from all electronics in. I was tring to remove the couterweights off the vibrator motors. I ended up destroying one of the motors, lucky i got spares. I read on the web that using a vise grip works. Anyone have any good sugestions to get rid of the counter weights.

What's up all, this is my first diary entry. I just found out about robots.net today by surfing the web. I think it is great that a community of robot builders exists and allows for communication between them. I've been working with robots for about a year now. I have built a working micromouse that has won several times. My next project is a tiny robot, solar powered, 2 pager motors, 3 solar sensors, 1 temp sensor, couple ir sensors, a little buzzer. I'm still looking for a small fast low power microprocesser. thinking about an arm or maybe just using a PLA for logic. I'm on AIM SN: 'sashi ono' give me a IM if you are interested giving me some new ideas on my new project.

Share this page