First, Merry Christmas to all. One of my latest purchases is the formula flowed buggy. I am evaluating it for college purchase. I have written a simple program where is should just move around in a square. When I simulate it in flowcode on the dashboard panel it does what I would expect. However, in real life it does a 360 degree turn at each corner, could anyone suggest why please?
Pete
Formula Flowcode issue.
Moderator: Benj
-
peterhalloway
- Posts: 30
- Joined: Sat Dec 29, 2012 4:49 pm
- Has thanked: 9 times
- Been thanked: 5 times
- Benj
- Matrix Staff
- Posts: 15312
- Joined: Mon Oct 16, 2006 10:48 am
- Location: Matrix TS Ltd
- Has thanked: 4803 times
- Been thanked: 4314 times
- Contact:
Re: Formula Flowcode issue.
Hi Pete,
When I did robotics at University the first lesson involved getting a robot like the FF to draw letters with a pen such as L or T. Marks were given for angles and lengths of lines. At the end of the lesson it turns out that dead reckoning is ok short term if you can play with your exact robot at that exact battery level and tweak the firmware speeds and delays to suit. Switching to another identical robot the letters came out very different due to minor differences in the motors and gearchain. This was a good learning exercise. What made things worse was we were using Lego so you had to give the magic squeeze to get it to work right for a demonstration.
This then leads on to using the on-board sensors to help navigate the robot in the real world. i.e. line following or wall following.
The simulation does not have the same problems in that there is no variable battery voltage and the motors will turn under a power of 1. In practise the motors don't actually begin to turn until you are at power 50-80 to overcome the moment of inertia. We have however tried to get things as close as possible with line following and maze solving programs working well for both without changing the code.
http://www.flowcode.co.uk/wiki/index.php ... ec2007b406
[/youtube]
When I did robotics at University the first lesson involved getting a robot like the FF to draw letters with a pen such as L or T. Marks were given for angles and lengths of lines. At the end of the lesson it turns out that dead reckoning is ok short term if you can play with your exact robot at that exact battery level and tweak the firmware speeds and delays to suit. Switching to another identical robot the letters came out very different due to minor differences in the motors and gearchain. This was a good learning exercise. What made things worse was we were using Lego so you had to give the magic squeeze to get it to work right for a demonstration.
This then leads on to using the on-board sensors to help navigate the robot in the real world. i.e. line following or wall following.
The simulation does not have the same problems in that there is no variable battery voltage and the motors will turn under a power of 1. In practise the motors don't actually begin to turn until you are at power 50-80 to overcome the moment of inertia. We have however tried to get things as close as possible with line following and maze solving programs working well for both without changing the code.
http://www.flowcode.co.uk/wiki/index.php ... ec2007b406
[/youtube]
Regards Ben Rowland - MatrixTSL
Flowcode Product Page - Flowcode Help Wiki - Flowcode Examples - Flowcode Blog - Flowcode Course - My YouTube Channel
Flowcode Product Page - Flowcode Help Wiki - Flowcode Examples - Flowcode Blog - Flowcode Course - My YouTube Channel