Questions regarding thesis of Zumi Robot

  • Hello Robolink Team,

    my name is Michael, I´m 26 and writing my thesis about Zumi, as it is going to be used at my university in Offenburg, Germany for teaching AI next semester.

    I spend some time with Zumi already and had some minor advices and many questions which came up in the process of working with Zumi, which probably only you can answer.

    I really hope, that someone who has a little bit of knowledge about the Zumi project is able to see into this message and help me out with the challanges i have been coming up against and is interested in 1 or 2 improvement suggestions.

    I´ve accumalted quite some questions, but i would be very thanksful if you find the time to answer all of them, as it would help me very much in my thesis.

    I´m always open for a call as well.

    Improvement suggestions:

    1. In Lesson 4.1 in the text is written, that you are supposed to show the final image with vision.show_image, but it always says, that vision doesnt have a function called show_image. I think you misstyped and ment to write that you show the image with camera.show_image()

    2. You should either have at least alle the functions used in the lessons in the Zumi Libary which is accessable online or write more details about the functions used in the lesson. I had at least 2 lessons I had trouble with, because I didnt know which parameter are allowed or even more import, what the functions return, which would have really been helpful.

    3. For example in the bottom IR sensor Lesson you simply said wirte Pseudocode, then write the real code, without explaining very much about how its done, which wasn´t really helpful i have to say. Bit more explaination would be great, since i had no idea how to make Zumi stop infront of a white line instead of on top of one, like it was shown in the picture.

    Points mentioned above are really ment constructive, to improve the Zumi project not to nag about things 🙂


    1. Is there a Libary or a way to see all the Functions Zumi has avaible? In the lessons alone there have been so many functions, which haven´t been noted in the Libary and I´m certain, that there are a low more.

    2. Our university also bought the big Roadmap, with the horizonzal white stripes on the road and locations like "school", "museum"... It was also briefly mentioned that it is possible to direcly navigate Zumi to one of that locations, I would like to know, how that is done.

    3. Is there any code, which you made or would make public of for example code you used for stuff you showed at conventions, like the Smile or look grumpy and Zumi reacts, or that you draw an object Zumi knows and he reacognizes it?

    4. How does Zumi follow the road on the above mentioned Roadmap?

    5. Are the methods to detect more that just smiles? Like look grumpy or other facial expressions?

    6. Where are the printable sheets which are beeing talked about in the QR-Lessons?

    7. Is there a way to activate autocomplete, so that when you type for example zumi. it shows you zumi.forward(), zumi.reverse, etc.?

    8. Is there more detailed technical data about the parts of Zumi. Like the frequency in which the IR can messure or more data about the camera , etc.?

    9. I struggled to make my own "I smile, you smile" Code work, which has been in the lessons und I´m not sure, wheater the function just isn´t working properly or if my code is wrong. Do you have a working code for that?

    10. I have been able to teach Zumi colors in the past, which apparently works diffrently now. When will the machine learning feature be useable again?

    11. In the Lesson for the bottom IR sensor is show and asked for, that you program Zumi, that he stops before he reaches a white line. I would like to know how that is done with the bottom IR sensors, as for me he only over stopped, when he was right on or past a white line?

    I know it´s a lot, but I would be really grateful to recieve answers, to that questions 🙂

    Kind regards

  • administrators

    @eggy Thanks for all the great questions! I think it would be best to answer them over a video call. I will contact you via email with a link to schedule one. Thank you!

Log in to reply

Looks like your connection to Robolink community was lost, please wait while we try to reconnect.