Free vocabulary hub
Sensors, actuators, and how machines learn from examples
Beginner robotics connects sensing, conditional response, and feedback. Machine learning and classification are introduced as pattern-finding from data—with attention to narrow or messy training data.
Measures a property of the world (distance, light, temperature, motion) and turns it into a signal a machine can use. Sensors are how robots “notice” their surroundings.
Real-world extension: Sensor systems are central in environmental monitoring, autonomous vehicles, and industrial automation.
Sources
Turns energy into physical action—spinning a motor, moving a wheel, opening a gripper. If sensors help a robot sense, actuators help it do.
Real-world extension: Actuators appear in robot arms, vehicle steering, prosthetics, and spacecraft mechanisms.
Sources
Using sensor information to adjust action while the robot is still operating—so it can follow a line, avoid obstacles, or correct drift instead of blindly continuing.
Real-world extension: Feedback control supports warehouse robots, self-balancing machines, and aircraft stabilization.
Sources
A program makes one choice if a condition is true and a different choice if it is false—what lets machines react rather than only repeat fixed steps.
Real-world extension: Conditionals are a building block for games, apps, traffic systems, and autonomous behavior trees.
Sources
Computer systems that improve from data. Rather than writing every rule directly, programmers let the system detect patterns that help it make better predictions.
Real-world extension: Used in recommendation systems, speech recognition, sorting systems, and scientific image analysis.
Sources
The set of examples a machine-learning system studies while learning a task. If the examples are narrow, messy, or misleading, the resulting model can be too.
Real-world extension: Careful dataset design matters in medicine, self-driving systems, and facial-recognition research.
Sources
Predicting which category an input belongs to—many beginner AI activities use classification because outputs are easy to see and compare.
Real-world extension: Object classification supports recycling machines, agricultural robots, and camera-based inspection.
Sources
Helps machines interpret images and video—detecting objects, estimating depth, recognizing terrain, or tracking a target.
Real-world extension: Used in autonomous driving, medical imaging, airport safety tools, and planetary navigation.
Sources
A machine can sense, decide, and act with limited direct human control. In practice, autonomous systems still depend on sensors, planning, uncertainty handling, and safety rules.
Real-world extension: A major goal in Mars rovers, sea-surface vehicles, underground robots, and industrial mobile robots.
Sources