Thursday, 15 November 2018

vislearn setup

Sneak peak: automatic position learning using vision

Normally with RoboForth positions are programmed using a teach pad. In JOG mode the user moves the robot in increments over and then down onto the object to be picked up, maybe making small adjustments in X,Y,Z and gripper angle then clicks the learn button on screen.

ST Robotics have been working on using a camera to locate an object and send coordinates to the robot controller. This includes the correct angle for the gripper. The software then learns the position as a PLACE name. The actual coordinates of the object are not of interest. The system can return to the learned position under program control.
If the position changes it may be quickly re-learned using the same vision system.

The process of learning the position involves moving the robot with camera over the object until it is in the viewport of the vision system. Rather than use the teach pad a joystick is employed which allows for continuous motion in straight lines in X,Y and Z directions.

The picture above shows the development setup. Please see the video here:

Ultimately it will be possible to learn many different positions and have different objects.

Monday, 5 February 2018

Talk to your robot!

Control of an ST robot is a man-machine conversation so why not use speech?

In reality you really need text to program anything complex but once programmed you could easily have a few voice commands. The speech engine we use is like the macro buttons on RobWin. You can say a single word but have the system send a whole string to the robot. Do see it in action:

Wednesday, 6 December 2017

We have created an art form out of an R12 robot. It has no works inside but red and blue LEDs. Then we put it in a glass cabinet in the reception in Orwell House (Cambridge).

Sunday, 5 March 2017

We now have JERK in the software. Jerk is basically the rate of increase of acceleration (like acceleration is the rate of increase of speed). To better understand it imagine you are in a car which suddenly accelerates. At the start you will push back in the seat so your personal acceleration will be delayed. In a robot this is when all the elasticity in the drive is taken out before actual acceleration can begin. Its worst manifestation is when the robot stops e.g. to grip something. Deceleration stops but the robot moves on before bouncing back. So with JERK this deceleration tapers off resulting in a smooth stop. The other new feature is you can program a special deceleration for emergency stop. We are not interested in precision in this case, we just want to stop as soon as possible. All this is now version 17 of RoboForth. I have developed the finest robot software on the planet and that's it. No more.

Wednesday, 16 November 2016

Our two students, Kazuo and Aaryaman this year did an amazing job creating a robotic dog. They both achieved Nuffield Research Placements scheme and the British Science Association Gold CREST awards.

A robot dog may sound simple but it was a major engineering exercise requiring CAD mechanical design, electronics and software including some quite difficult problems for example, synchronism of leg motors. What was especially satisfying was seeing them collaborate often disagreeing yet always talking through problems and agreeing a solution. When they encountered a problem I would do no more than suggest a solution or a particular component or give them the tools they needed (e.g. digital oscilloscope). They would then dive in and leave me behind with their thinking. They were creative and true problem solving engineers. They will go far.

Oh and guess what they named it: K9 !

Friday, 7 October 2016

wow, really old video of Joseph Engelberger and one of the first industrial robots on the Johnny Carson show.

Saturday, 9 July 2016

15th June 2016:
Another successful day at Duxford showing the students how to build and program robots
Our people left to right: Kyle, 'DJ', Kutlo.