Self-driving car: school robotics curriculum?

Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 20,866
edited March 31 in Robotics Vote Up0Vote Down
I've taken a year off from teaching robotics at the local high school, and I'm trying to decide whether to return next school year (assuming they want me back). The problem with the course I've taught is that kids can take it as many times as they want, and many do return. So the curriculum can never be the same as it was the year before. Thus, I will need to come up with something completely new.

Self-driving cars are a huge deal these days. Even the local school's Mock Trial kids have had to argue a criminal case this year involving a self-driving car: the Auto Auto. So I've been thinking about how to create a robotics curriculum around this theme. It would have to use the ActivityBot, since the school already has plenty of them. Also, there's a gymnasium/basketball court adjacent to the robotics classroom, where the navigation could take place. So here are my questions to the forum:

1. How would you configure a challenging -- but navigable -- layout?
2. What visual cues should the kids' robots be able to deal with?
3. What sensors would be necessary to navigate it successfully?
4. What objectives should the students be able to master?
5. What dynamic challenges could be introduced to test accident avoidance?
6. Or is such a project too ambitious -- for the kids or for the available hardware?

Hopefully this thread will attract other instructors, and maybe we can come up with a curriculum together, if such a thing is even feasible.

Thanks,
-Phil
“Perfection is achieved not when there is nothing more to add, but when there is nothing left to take away. -Antoine de Saint-Exupery

Comments

  • 5 Comments sorted by Date Added Votes
  • 1. since it should simulate a self driving CAR you could provide some streets and parking places. Maybe a small version of the actual school?
    2. Signal lights and crossings come to my mind, if no signal lights then at least 4-way-stops? Traffic jams for sure. Has to be realistic.
    3. Servo mounted Ping should be able to detect other cars on crossings. Signal lights might be expensive to detect.
    4. Stay on the roads, avoid collisions with other cars, figure out a way to find empty spot on a parking lot and park in? Finding shortest routes?
    5. Pet animals? Simulated persons using a crosswalk?
    6. A line follower is much alike a self driving car, but that is getting old, slowly. So bumping it up with more challenges seems OK to me.

    Actual self driving cars use GPS and to a big extend roadmaps. So providing all students with some sort of premade map on SDcard for the planned layout could make it easier.

    Extra challenge, one group (or teacher) could build drunk drivers, making the course more dangerous?

    Enjoy!

    Mike
    I am just another Code Monkey.

    A determined coder can write COBOL programs in any language. -- Author unknown.

    The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this post are to be interpreted as described in RFC 2119.
  • I like to think about this stuff. I asked the guy that runs the self-driving car meetup here if he has anything, he usually responds pretty quick. In the meantime here are my thoughts.

    1. How would you configure a challenging -- but navigable -- layout?
    Use black tape for roadway edges, and setup line following so it works on the edges instead of the middle. Have a couple of intersections. Maybe a parallel parking spot or LED traffic signal. See how many bots can drive around before gridlock.

    2. What visual cues should the kids' robots be able to deal with?
    High contrast edges to keep it on the road, PING to avoid hitting another vehicle.

    3. What sensors would be necessary to navigate it successfully?
    QTI and PING, encoders wouldn't hurt.

    4. What objectives should the students be able to master?
    Keeping traffic flowing.

    5. What dynamic challenges could be introduced to test accident avoidance?
    Select bots could have a random chance that it will not follow certain rules, such as following distance.

    6. Or is such a project too ambitious -- for the kids or for the available hardware?
    Can't be more ambitious than the sailboat, that was awesome.

    Whoops, I forgot about PINGs interfering with each other. Maybe that could be part of what they have to overcome.
  • Some good suggestions so far. For the roadway I would suggest using rubber sheeting like the pond liner used for water gardens. Cut it to the desired width and shape to make straight and curved roads and intersections. Makes it easy to set up and remove it and allows for the layout to be varied.
    In science there is no authority. There is only experiment.
    Life is unpredictable. Eat dessert first.
  • ercoerco Posts: 17,674
    I saw "Ghost in the Shell" yesterday, good flick. Curiously, not a single self-driving car in the movie. But in the trailer for the upcoming "Fate of the Furious", a beautiful evil woman takes over every self-driving car in the city and goes on a rampage.

    Let the good times roll.
    "When you make a thing, a thing that is new, it is so complicated making it that it is bound to be ugly. But those that make it after you, they don’t have to worry about making it. And they can make it pretty, and so everybody can like it when others make it after you."

    - Pablo Picasso
  • xanaduxanadu Posts: 2,849
    edited April 1 Vote Up0Vote Down
    The guy that does the self-driving car meetups said they're following this guide. It was made for Arduino, but the 6 objectives would apply. I think your returning students might be beyond some of it.

    I mentions accelerometer, augmenting self-driving with tilt sensing you could have a hill climb, or something besides flat roads. Or a flat road with a ditch on either side.
Sign In or Register to comment.