Engineer Boldly: AI Robotics Sprint

Frequently asked questions

This robot looks like a toy — what use would this be in a factory or real setting?

Its an educational robot and our mobile platform is deliberately minimal: compact, open-source, and expandable. But under the hood, you're running the same architecture used in companies like Waymo, Boston Dynamics and NVIDIA — ROS2, Linux, SLAM, vision transformer models, reinforcement learning, real-time sensor fusion and onboard AI. You can also add to your robot as you want to learn new things, for example add a manipulator arm to learn Inverse Kinematics, or add other sensors and training data collection mechanisms. The lessons you’ll learn with this robot are directly transferable to the cutting edge in AGVs, robotic arms, and vision-guided factory systems. We chose this platform not because it’s flashy — but because it’s transparent, hackable, low cost, and built to teach the skills that actually matter in real-world deployment: perception tuning, dataset collection, model training, debugging, and iterative autonomy.

How technically difficult do you expect it to be?

This course is designed to challenge ambitious students with real-world robotics and machine learning. You’ll implement AI perception systems on a mobile rover platform with two degrees of freedom. The math stays within high school-level concepts—algebra, geometry, trigonometry, and some matrix notation from linear algebra. Prerequisites include familiarity with Python and basic comfort using a Linux command line. Expect to be technically challenged, learn independently, and come away with practical fluency in ROS2, machine learning, computer vision, and robotics. It’s not a beginner’s camp—but it is open to anyone ready to dig in, do the work, and build something real.

What will the schedule be like on Saturday, how much work in between?

We expect that our students will have full time commitments outside of our course and so it really depends on each individual and how much time they want to dedicate. On Saturdays we will have lecture in the morning and then student build/experiment time in the afternoon. The student work time can be more flexible, we will make the space available and have office hours at some other times during the week based on interest.

Do I need to know how to solder or use tools to build my robot?

No soldering needed—we take care of that. But yes, you’ll be connecting wires, mounting cameras, handling batteries, and tightening screws and bolts. It’s all part of working with real robots. Things come loose. Parts shift. Every time you power it on, expect to do a little hands-on maintenance. That’s not a flaw—it’s the point and that's what we are here to help with. Learning how your robot is built, how to fix it, and how to keep it running is core to becoming a real robotics engineer. Simulations don’t teach that.

Can you tell me more about the robot kit and what it will look like?

We’re developing an open-source mobile base with sensors as the core kit. The design is still evolving, so parts may change, but the goal is a sturdy, AI-ready robot you’ll be able to build, teach, debug, and expand. You can follow the latest bill of materials (BOM) and design notes in our GitHub repository https://github.com/RoseCityRobotics/hardware-builds/blob/main/boms/mobile-base-with-sensors.md