How it all started...
This is the first post, but we've actually been building our robot for about a year now...
Our team comprises Benjamin (15), Jeremy (13), Daniel (6), Jenny (Mum), Kok-Fu (Dad) and John (‘The Godfather’).
Experiments began in September 2018, I've attached a few highlights:
We haven’t finalised the design yet but will be building our robot out of Lego because it is versatile and the children are already very familiar with making their own Lego creations. We have experimented with different drive trains including conventional ‘rc car’ approaches with steering wheels, suspension and differentials. We have also looked at tracked designs and designs using fixed independently driven wheels (see photos).
Our preference is for a tracked design with each track independently driven. It avoids the need for suspension and complex steering. Our plan is to use Lego power function Motors to provide drive, but connected to drive boards controlled by an Arduino controller. Arduino is very simple to program, but if we can get the Pi to directly interface to the motor driver board we may dispense with the Arduino.
The intention is to use a Raspberry Pi 3B+ to provide intelligent function, but we may switch to Pi Zero if we can get away with less processing (we have a concern about the power consumption of the 3B). The tracked chassis will have a line follower sensor and ultrasonic distance measurement, but we are thinking about supplementing with camera based machine vision (OpenMV) for the Autonomous challenges. This will be the most difficult to do, but ‘Uncle John’ has offered to help out. He is based in Canada, so it will be interesting co-ordinating the development effort! Welcome any ideas on tooling to help do this practically. Ideally we want to be able to tweak code from either side of the Atlantic and to easily upload into Pi to observe what happens and to debug.
Danny, our youngest team member (aged 4) likes Wall-E, so we think the basic design will look similar with tracked chassis and a body with a camera based machine vision ‘head’. We may even make it ‘talk’. We are going to call it Dan-ED. I’ve attached some photos of what we have been experimenting with so far.
‘RC Approach’, with steering wheels,
suspension and differentials. (Too
complicated!)
Fixed independently driven wheels
A tracked design with independent drive
for each track (favoured approach).
Our team comprises Benjamin (15), Jeremy (13), Daniel (6), Jenny (Mum), Kok-Fu (Dad) and John (‘The Godfather’).
Experiments began in September 2018, I've attached a few highlights:
We haven’t finalised the design yet but will be building our robot out of Lego because it is versatile and the children are already very familiar with making their own Lego creations. We have experimented with different drive trains including conventional ‘rc car’ approaches with steering wheels, suspension and differentials. We have also looked at tracked designs and designs using fixed independently driven wheels (see photos).
Our preference is for a tracked design with each track independently driven. It avoids the need for suspension and complex steering. Our plan is to use Lego power function Motors to provide drive, but connected to drive boards controlled by an Arduino controller. Arduino is very simple to program, but if we can get the Pi to directly interface to the motor driver board we may dispense with the Arduino.
The intention is to use a Raspberry Pi 3B+ to provide intelligent function, but we may switch to Pi Zero if we can get away with less processing (we have a concern about the power consumption of the 3B). The tracked chassis will have a line follower sensor and ultrasonic distance measurement, but we are thinking about supplementing with camera based machine vision (OpenMV) for the Autonomous challenges. This will be the most difficult to do, but ‘Uncle John’ has offered to help out. He is based in Canada, so it will be interesting co-ordinating the development effort! Welcome any ideas on tooling to help do this practically. Ideally we want to be able to tweak code from either side of the Atlantic and to easily upload into Pi to observe what happens and to debug.
Danny, our youngest team member (aged 4) likes Wall-E, so we think the basic design will look similar with tracked chassis and a body with a camera based machine vision ‘head’. We may even make it ‘talk’. We are going to call it Dan-ED. I’ve attached some photos of what we have been experimenting with so far.
Our experiments with Arduino and
motor driver to control standard robot chassis.
We will use Pi 3B+ for overall control.
Comments
Post a Comment