School robots - robots for educational institutions
This article is intended to provide an initial overview for people from schools (not elementary schools!) and inter-company educational institutions. It assumes little to no knowledge of robotics and different budgets. The author has been observing the robotics industry for years, but is not an expert on school-based robots. Nevertheless, he sees the declining interest in STEM in Germany. Perhaps robotics in the classroom can help counteract this. The topics of financing robots as well as the possibility of funding for inter-company educational achievements will then be addressed.
Basic
Robots have been used for decades. The classic industrial robots, as we know them from the automotive industry, are very fast, precise, often strong, but also inflexible and too dangerous for schoolchildren without safety measures. Cobots, on the other hand, have been developed to work with humans. These are lightweight robots that are generally easier to program and sometimes even perceive humans thanks to sensors. Nevertheless, they can also pose a danger, e.g. if they move something pointed at eye level. However, they are not suitable for many schools, especially because of their price. Established models quickly cost upwards of €20,000. For a school class with at least five robots, 100,000 € would have to be calculated in. Smaller robots, some of which are even available with educational packages and which function in the same way as the large productive robots, therefore seem to make more sense.
Among the important basics for "robotics newcomers" should be the knowledge of the essential distinguishing features.
- Payload: This indicates how much a robot arm can lift. Important: The weight for the related gripper must be subtracted from the payload. (By the way, robots almost always have only one arm).
- Range: The term is self-explanatory (it indicates up to what distance a robot can grab something from its base). Important: The range indicates the maximum value and does not take into account the mobility. This depends on the number of joints/line of freedom. If you imagine the forearm as a gripper, i.e. from the elbow to the wrist, the reach might be 40 cm. But something at a distance of 20 cm - i.e. within the reach - could hardly be grasped, because the forearm is rigid. The whole arm of the human being including the joint at the shoulder and the wrist, on the other hand, can grasp almost everything within its reach. It is articulated. Typical large robots therefore have 6 lines of freedom. To mimic humans, they tend to need 7. Educational robots have fewer for cost reasons, but there are differences among them as well.
- Speed: This does not play a significant role in the school. There are no cycle times, etc.
- Repeatability: This value indicates possible deviations and can be more significant. If, for example, typing on a keyboard is to be taught using a robot, a repeatability of 2 mm would be unsatisfactory. Probably, the wrong key would be hit too often. This criterion is therefore more important.
- IP protection class: What kind of pollution can a robot tolerate? Water-accepting robots, for example, have IP 67. This is irrelevant for school, but not for some industrial activities. For school, even a simple IP 20 class should be sufficient.
- Programming: How is the robot programmed? Here I ask for own research if the budget stands. If there is no money, you take what is available. If there is enough money, it is worthwhile to compare (programming language, etc.) For industrial robots, the programming is a big issue because it is very complex. Educational robots are usually much easier to handle.
- Accessories: It is helpful if some interfaces/ports are available to connect e.g. a webcam etc.
- Stability: In my opinion, this is an important criterion, especially if the small robots are not permanently mounted. In this case, they should also be able to withstand knocks when walking past, etc. without falling over immediately. Depending on the model, this can be helped with a screw clamp.
| Dobot | Niryo | Elephant Robotics | |
|---|---|---|---|
| Magician | Ned 2 | myCobot 280 | |
| Axes | 4 | 6 | 6 |
| Reach in mm | 320 | 440 | 280 |
| Load capacity in g | 500 | 300 | 250 |
| Repeatability in mm | 0.2 | 0.5 | |
| Dead weight in kg | 3.4 | 7.0 | |
| Cost approx gross | 1,500 | 4,000 | 800 |
| Link | https://variobotic.de/robotik-in-schulen/dobot-magician/dobot-magician-technische-daten/ | https://niryo.com/products-cobots/robot-ned-2/ | https://www.elephantrobotics.com/en/mycobot-280-pi-2023-en |
Now some models that seem affordable even in larger numbers.
School robots and educational robots from Dobot
Variobotic from Ulm has not only been offering robots from the Chinese manufacturer Dobot for several years, but also service, educational material and other accessories for using the robot correctly. The company is a member of the German Robotics Association and, in my opinion, employs a trained pedagogue who is exclusively concerned with educational issues. The smaller robots offered by the manufacturer are - permanently - so permissible that they are also used by numerous industrial customers and there in large numbers when it comes to smaller payloads and shorter reach. The low price starting from 1,500 € (gross!) is explained by the fact that these are not "shrunk" variants of the well-known industrial robots (i.e. one arm with 6 axes), but the mechanics comprises only four axes. This is of no relevance for programming or concrete applications. However, the robots are not quite as mobile. For school purposes, however, this seems negligible.
In addition to 3D printing (!), classic robot tasks such as optical recognition (vision), vacuum gripping, moving on a linear axis or gripping from a conveyor belt can also be learned and tried out. All this hardware can be ordered from the same supplier and thus suitably. This of course eliminates other expenses, but frugal schools can mix and match. I.e. buy e.g. 10 robots in addition in each case 3 conveyor belts, 3 x vision etc.. The pupils learn then evenly somewhat shifted. I.e. the equipment of a classroom with a budget of approximately 25,000 to 30,000 € seems to me well possible.
Niryo: Similar to before, but more expensive and from France
OrangeApps
While I would recommend the first two solutions - as an observer from the sidelines - I am not sure about the OrangeApps solution. On the one hand, it consists of 850 Lego bricks, so I would judge it as fragile. Then the solution is already several years old and the whole system is very KUKA-heavy. KUKA builds very good robots, but still uses complex software. Since this is rejected even by many medium-sized companies, KUKA is hardly represented in this market segment. One of the plus points is the low price of 1,190 euros. For this, however, the robot must be assembled. After all, it has 6 axes and is thus very mobile.
The teach panel (tablet) shown in the video is not included in the price and will probably quickly cost €1,000, as it is from KUKA.
Elephant Robotics
Another "shaky candidate" are the robots of this Chinese manufacturer. Since he does not have a branch in Germany or the EU, the risk increases with any warranties. But it has been around for years, is very inexpensive and has a large range of accessories. Also in programming it seems very interesting (incl. ROS) and also allows the use of console controllers. Cf. here LInk. In addition, it can be mounted on a small mobile robot, which costs about 900 US-$. While the model "MyCobot280" can lift only 250 g and costs just 800 US-$, "MyCobot320" can lift 1 kg, has various other features (JavaScript), but costs 2.400 US-$.
If you are interested, maybe contact me. Elephant Robotics is always looking for contact with me. Therefore I could act as a mediator. In case of problems, Elephant might be more accommodating to avoid negative reporting. Maybe together we can get a cheaper test robot. I.e. you test and keep it and report on this website in return.
Professional robotics sets for educational institutions
Individual suppliers have supplemented cobots, i.e. lightweight robots that are also used in companies, in such a way that they are ideally suited for training purposes. These learning stations are intended for inter-company training centers or vocational schools/training workshops. Since the basic version of the robot used here already costs upwards of €20,000, such a package is likely to quickly cost €30,000 +.
The Glaub company is active in northern Germany.
Jugard+Künstner is more active in southern and eastern Germany.
Alternatively, you can assemble your own learning station. In this case, it is highly recommended to take a look at the Franka Emika Cobot, which comes from Munich and is manufactured by TQ in the Allgäu region. The research version is quite inexpensive. The Deutsches Museum recently purchased 128 Franka Emika for its AI factory. The research institutes of the Technical University of Munich also prefer to use it.
Funding for secondary schools
If there is no budget available and the sponsoring association can't afford everything, I strongly recommend contacting local companies. Nowadays, many companies spend a lot of money on training fairs, so a little presence at school can also be feasible. Unfortunately, I can't give more helpful tips to secondary schools.
Funding for inter-company training centers
There is a generous program from the Bafa. 90% of the expenses up to 500,000 € can be subsidized(link). I will gladly accompany you on this complex application process and also in the selection of the robots.
Are we networking? LinkedIn
-> Go to the Cobot group on LinkedIn (link). The author is also a consultant (robotics, tech & finance). Hardly anyone is likely to have a comparable market overview.
