Course 3 of 5 · Free
Control and Balance
Feedback, staying upright, following a line, and finding a way through.
20 lessonsabout 20 min in totalno account needed
After this course you can
Describe a feedback loop and why a robot wobbles without one.
Start here
Feedback Control Explained
Measure, compare with the goal, correct, repeat. Cruise control and robots use the same loop.
Start the first lesson →One minute. No account needed.
The lessons, in order
- 1Feedback Control Explained Measure, compare with the goal, correct, repeat. Cruise control and robots use the same loop.1 min
- 2Line Follower Robot Two light sensors see dark line or bright floor. If the line drifts to one side, the robot steers toward it.1 min
- 3Why You Don't Fall in a Moving Bus (And Robots Do) Your inner ear and muscles correct your balance constantly without you noticing. A robot must be built to do the same.1 min
- 4How Boston Dynamics Robots Keep Their Balance Legged robots sense their body tilt and joint positions many times per second and keep shifting weight and placing feet to stay up.1 min
- 5How Drones Stay Level in Strong Wind A drone measures its tilt hundreds of times a second and changes each propeller speed to push back against the wind.1 min
- 6PID Controller in 60 Seconds A PID controller corrects an error using three ideas: how big it is now, how long it has lasted, and how fast it is changing.1 min
- 7ROS 2 ROS 2 is not a real operating system. It is a toolkit that lets many robot programs talk to each other by sending messages.1 min
- 8How Robots Learn to Walk The robot tries moves, gets a score for good results and learns by trial and error, mostly in simulation first.1 min
- 9A* Pathfinding A* explores the map by always trying the most promising square first, using the cost so far plus a guess of the cost left.1 min
- 10State Machine A robot is always in exactly one mode, and a few simple rules decide when it switches to another. That is a state machine.1 min
- 11Robot Latency Add a blink of delay to the loop and a balancing robot goes from steady to wobbling. Latency is how late the robot acts on what it sensed.1 min
- 12Reward Hacking A learning robot found a way to score more by never finishing. Reward hacking is when the score and the real goal come apart.1 min
- 13Sim to Real A robot can fall a million times inside a computer without breaking anything. The hard part is making what it learned work on a real floor.1 min
- 14How Humanoid Robots Walk Without Falling (ZMP Explained) A walking robot is always almost falling. The zero moment point is the idea that tells it where it can safely put its weight.1 min
- 15Gain and Damping Correct too hard and the arm shakes. Correct too gently and it is slow. Gain sets the push, damping calms the overshoot.1 min
- 16How Self-Driving Cars Think A self-driving car does more than see. It also guesses what happens next. Perceive, predict, plan is the three-step loop behind it.1 min
- 17Swarm Robotics No ant knows the plan, yet the colony finds the shortest route. Swarm robots use simple local rules to get the same kind of result.1 min
- 18Robot Dog (Quadruped) Explained Four legs give a robot a stable base, and a clever walking pattern keeps it upright on rough ground.1 min
- 19How a Drone Hovers Four spinning propellers, tiny speed changes, and a sensor that checks hundreds of times a second keep a drone hovering.1 min
- 20Behavior Tree A behavior tree organises a robot's choices as branches of small tasks. When one option fails, the tree simply tries the next, so a Plan B is built in.1 min
Get one robotics email a week
Plain English. Free. Unsubscribe in one click.
Going deeper: every lesson links into the Atlas, and the free course path covers the same ground in longer videos with quizzes and certificates.