Showing posts with label Pushing robots. Show all posts
Showing posts with label Pushing robots. Show all posts

Thursday, 12 January 2017

Pushing robots


The purpose of this article is to explain the technical choices of the Cooperative robots project (the general article was posted on Wednesday, 23 November 2016 on this blog). According to the first article, the goal of this project was to make 5 sumo robots move together, in front of the main robot, the RICA.


To realize this project, we had at our disposal some specific material :
Firstly, the base of the robot was a Pololu Zumo robot. With this robot, all the parts needed for the locomotion (caterpillars, motors, power source and motor management) were ready to use.

Afficher l'image d'origine

To reach the goal of this project, one of the main challenges was to be able to detect the other robots. To do that, we could use three types of sensors, numeric infrared, analog infrared and analog ultrasonic.
For the detection of the other robots, most of us chose to use analog infrared sensors, because it is easy to use and provides distance informations. Some of us also use ultrasonic sensors to detect obstacles like walls
.
Afficher l'image d'origine


The other aspect of the detection was to make the robot easy to detect by the others, so it was very important to work on the shape of the robot. We choose to build an outer shell with flat sides to make a good surface for sensors. This external shell was made with a 3D printer, because it was an easy way for us to build lightweight parts.  




Finally, to gather the information from sensors, execute the program and command motors, we used a microcontroller Arduino uno. In an upcoming article we’ll see the coding steps.

Afficher l'image d'origine

Wednesday, 23 November 2016

Cooperative robots



“We all do better when we work together. Our differences do matter, but our common humanity matters more.”

― Bill Clinton





We will put aside the humanity for now but, the rest of this quote pictures well the subject of this article. As we said in an older post, blind cells are very old rooms. Since their construction, a lot of scientific experimentations took place here over the years. The result of this is that the ground of the cells are filled with a lot of garbage and contaminated dust.  Even for a tank robot it's hard to move in this kind of environment and it can also cause contamination.

So an important need is highlighted here and it is to clean the way before the dismantling robot RICA. Implementing this would avoid the potential surface contamination of the robot. It will improve the survival of  electrical components and simplify the clean-up process of the robot.

One solution to answer this need is to add some pushing robots that will have to push the objects out of RICA’s path. Those little robots will have to collaborate with each others to help the RICA to move properly in the cell.

Therefore the realisation of those little robots was a subproject of the big project RICA IV that our class has to realise. During our first year of formation we had to create the first prototype of those cooperative robots. This first prototype was an academic project to make us apply our knowledge in programming and automation.

The specifications asks that five robots must be used, they're disposed in before the RICA in a row like we can see hereunder.

C:\Users\Paulo\AppData\Local\Microsoft\Windows\INetCache\Content.Word\Capture.png


The central robot has to control its position considering the RICA's, it has to stay before the RICA at a settled distance without losing the dismantling robot. Each of the four side robots is guided by the one who move before it, to their left for the robots located on the right and at their right for the ones located on the left. Each robot has to be able to work in any position of the row, so they have to work in different modes that can be switched easily.

The class was divided in 5 groups, each group had to create one robot, they had to work together and with the RICA.

Unfortunately we did finished this project with four robots instead of five because a group had Human Ressource problems. In the end all the four robots were working and were physically different, it illustrates the distinct points of view of each group.

The next step to improve those robots would be to set up a wireless communication between them and an artificial intelligence to get real autonomous robots that can organize each other without human help.


[soon : link to video of robots working together]
[soon : link to robots’ realization]