An apology, and I did get the robot to work

VEX V5 General Discussion Technical Discussion
1

TL;DR
I’m sorry for the vast posting confusion that I caused.

I live a charmed life, I have a great robotics team that I mentor, I have amazing parts to build robots out of, my life is good.

I posted "I want sensors post because I want sensors. I’m three deep in Sparkfun, Adafurit, Robosyssey etc. sites with sensors that I’d love to use not so much from VEX

I have a robot lab. You’ve seen/heard about Dean’s on “Lifestyles of the Robotic” and the tours of his 1.2 million sq ft lab. My lab is 12*12 and it looks like a Radio Shack and three ham fests barfed up their contents into the room. Cool part is that every part made in the last 10 years is only an arm-reach away, bad news is that EVERY part made in the last 10 years is within arm reach.

So plan on seeing me on “Hoaders – the robots!” in the next year on TLC.

I was working on part of a robot. I never build full robots, there isn’t the time. But I do work on partial assemblies, for example I’ve built the “far west coast” planetary drive. (It’s pretty cool) So I have my hand in building stuff.

I have the mecanum wheels from January and I built the basic squarebot. And it didn’t work (off track displacement) So I added the built in encoders and that started the spiral into disrepair.

Stop

I love Kevin Watson, I’m a PIC programmer from waaaaaaay back. Kevin published PIC code that make hard tasks easy. Part of his code became the WPI library. So you have highlevel (move robot) code, Kevin code on the details of a PID and the ability to bit bang the PIC processor.

Life is good

Then the Cortex came out. EasyC has a really painful GUI (I want type code) and RobotC (which my teams use) because we can type code, but its nnot bare metal like I can get to with EasyC with some workarounds.

Workarounds are the bane of my robotics…

So to make a longer story shorter, I’m in Muppetbot labs working on a simple mecanum wheel/motor/sensor and I reach the frustration point of the season with the IFI change, the RobotC change and me not being able to get to the metal.

So I posted into the now famous thread. And yes I know better. Yes I have lectured others on this in the past. And Disciple posted and while we don’t see eye to eye, I responded. And I know better and yes I have lectured others on this in the past.

And I heard my Mom say “Walk away”. (We all have Moms. listen to her) And I did.

I was / am pleased with the collection of mentors that wrote to me. Most were along the lines of (Foster WTF?) And that helped. And I went into the lab and got the code and the robot to work.

Takeaways for you that read this far:

Dean and jpearlman rock. Their simple 6 line posts of what goes on does not cover the decades of work they stuck into your problem to figure out what happens.

I’m not a role model. Karthik, JVN, Paul C, Rick K are role models. I want to be Rich K when I grow up. (I wanted to be JVN, but he lost his hair) I have a passion for kids (roboteers) and competition robotics, I want my roboteers to have access to the life I have (living in a Radio Shack closet :rolleyes: )

We can’t do this depending on others. I posted a thread about sensors, I got a response from “Lyncas” and others backing me up. We do amazing things, we can do more with better tools.

RobotC and EasyC are pointed at the schools. NASCAR is us, your Dad in the Ford is the schools. We need to find a way to make us all be better. While 2:20 isn’t a big deal, on the floor at Disney it’s a lifetime. I offer 8 matches to a team, 18 minutes for 4 weeks of build time. That is not really fair. So we do 12 events and that seems about right. So I still want a way to make it fun.

So to summarize a summary; Listen to you Mom. We can do better than your Dads Ford. Pick someone else to be a role model, Foster is just crazy about robots and kids. And remember it’s nice that you are important, is is more important that you are nice.

Thanks for your support!

Foster

2

:):):slight_smile: mmP (that’s supposed to be a thumb up).

3

Well put Foster :slight_smile:

4

Your work space sounds alot like my desk… :smiley: I had just cleaned it recently Proudly mounting my Telescoping arm on the wall.

In all seriousness very well put, you remain as a role model in my eyes.

  • Andrew
5

Foster;

I have been feeling a little, “under the weather”, this week so I have not summoned up the Energy to read your Two, Seminal[3] posts…

My first though was, “Yikes!!! Someone has totally gone Off the Rails…”, and then I realized that it was one of the most dedicated mentors I have read about… Then I realized, This must be really bad…

It sounds like things are going better, now…

I really respect the balance between “The Competition Platform of Vex” and “The Hobbyist Use of Vex”, but I see the Cortex Controller as so very much more Closed and less easy to “dig into”, than the PIC Controller. IFI/Vex Robotics has provided the very open, Vex Pro, with Open Source Tool-Chain, but is currently limited to Hobbyist and College Competitions only.

I too would like to see a Documented, Minimal Library for the Vex Cortex that could be interfaced from any Compiler targeting the Cortex M-3, C++, Ada, Forth, LOGO or whatever…

I would also love to see additional sensors, even if they’re 3rd Party, that would be allowed. This means that the Beginning Programmers, will be at a disadvantage of using more advanced Sensors, until support could be added to EasyC and RobotC, but for the Advanced Programmers, ( that might have an advantage ), a Handicapping System could be introduced to offset any Significant gains due to use of better Sensors or Extreme Advantages of Hardware.

I was really dismayed to learn that Team Repositioning of the Robot During Autonomous was allowed, because it seems there is no way to Achieve Results, completely Autonomously… It can not be the Processing Power of the Cortex, because I have seen FRC Robots with PIC Processors do fantastic things. So it comes down to the Allowed Sensors, and the Programming Skills of the Compiler Libraries and the Programmers themselves…

I also realize that Sensors like the CMU2CAM, are quite a lot of Money, but there seems to be many Sensors at the Robot Hobby Stores that could be used… If they were Certified to be Legal for usage…

The Master Code needs to be Frozen, at a point, so that Teams can test with it before a Competition. Beta Master Code can be released with new Sensors, if needed, and a Highly Optimized Feedback/Response and Trouble Shooting System put in place to get the Beta Master Code, to General Availability ( GA ) state.

Back on 3rd Party Sensors… Beta Master Code can be released for 3rd Party Sensors, or Libraries could be developed, and use the same Highly Optimized Feedback/Response and Trouble Shooting System put in place to support the Master Code updates, to insure optimal support and reliability of the sensors.

I also have a Big Nostalgic Streak…

I would like to still use the Cortex with the old 75Mhz controllers, or mix the Cortex with VEXnet, with the VEXnet Upgrade for the 75Mhz Controllers… Or allow the interfacing between the 75Mhz Controller as the Second Controller and the Cortex Controller on VEXnet.

I would like a Source Available, or at least a Non-Restricted License, Library, with Full Documentation so any Cortex-M compiler could be used to program the Cortex Controller. This would help with maintaining, The Nostalgic Support of Vex.

Every “system that grows”, goes through Growing Pains… I guess that Vex has reach one of those points.

6

Here was my desk at work, 04-SEP-2011. It’s a bit cleaner now… And needs to be cleaned again…

[ATTACH]5980[/ATTACH]

( Notice Vex 75Mhz Controller with VEXnet upgrade and VEXnet Key, lower right, under desk, on box )

Questions Answered about any Item(s) in picture…

( Zip File, Full Size )

HPIM4990.jpg
HPIM4990.jpg920×693 92.4 KB

HPIM4990.zip (1.82 MB)

7

Woah…

[offtopic]
What do you do as a day job? You, Quazar, and jpearman seem to know a little (oftentimes a lot) about everything digital…
[/offtopic]

//Andrew

8

[continuing off topic]

Digital, Large Diesel Engine Monitors and House System Control and Monitors… Focusing on the Recreational Vehicle Industry.

SilverLeaf Electronics Inc.

Here is a Device that I work with:

SilverLeaf TM-102

[ATTACH]5986[/ATTACH]

The TM102 monitors and controls a number of systems in the coach and is controlled with a VMS, HMS, or Omniscope diagnostic tool through the RV-C data network. The TM102 may monitor the tanks systems (fresh, gray, black, lpg), electrical systems (inverter, generator, transfer switch), and many other systems in the coach (tile heat, temperatures, awning, etc.) depending on its configuration. It also is the brain behind the autogenstart system. The TM102 can be configured through the Omniscope diagnostic tool.

There are 6 Analog Inputs, 12 Digital Inputs, 8 Digital Outputs, Two UART Serial Ports, Two CAN Bus Ports and an I2C port ( Sounds almost like a Cortex Controller )

Here is a Control Panel, the HMS062 that I did most of the Programming on, that communicates with the above TM-102, to display the Information received on the Analog and Digital Inputs, the Serial Ports, the CAN Bus Ports and the I2C Bus.

[ATTACH]5989[/ATTACH]

You learn some Electronics, you work at a place that Engineers, Manufactures and Trouble-Shoots Electronics, and when you get into Robotics, it’s mostly the same, just slightly different Sensors, and definitely different Controllers… But they all basically work the same…

[/continuing off topic]

HPIM5182.jpg
HPIM5182.jpg1010×761 91.3 KB

hms062.jpg

9

Foster,

There are some pretty complex sensors that most students don’t use in programming autonomous (http://www.vexrobotics.com/products/accessories/sensors/276-2333.html)
The Gyro helps with single axis movements (if other teams get in your way in autonomous)
(http://www.vexrobotics.com/products/accessories/sensors/276-2332.html)
The Accelerometer helps with the speed of your robot, and if you have a collision with another robot, etc.
These sensors are complex to use, and their not cheap. I dont understand why you need complex sensors though? Most teams use wait times, and can be very successful for them? I mean im not a programmer, so i probably wouldn’t understand. I do most of the designing and building, rather than programming and electronics. Hopefully this helped. :slight_smile:)