When Does a Machine Become a Robot?

A blender has a motor. A garage door opener has a controller. A thermostat has a sensor. None of those facts, by themselves, make a system a robot.

For this course, a useful working definition is narrower:

A robot is a physical system that can sense something about itself or its environment, use a controller to make or execute a decision, and act on the physical world.

That definition is not perfect. Robotics is a field with fuzzy edges. It is useful because it forces us to look for the system loop instead of arguing over whether something looks like a robot.

The pieces that matter

Consider a small mobile robot asked to stop 20 centimeters from a wall.

distance sensor
      ↓
measured distance
      ↓
controller
      ↓
motor command
      ↓
motor + wheels
      ↓
robot moves
      ↓
new distance to wall
      └────────────── back to sensor

The sensor does not "know" there is a wall. It produces a measurement. The controller does not "want" to stop. It executes logic. The motors do not "understand" the command. They convert electrical energy into mechanical motion.

The system looks intelligent only when those parts cooperate.

Classify the machine by evidence

Use the table below as a model.

The interesting case is the remote-control car. It has robotic hardware, but much of the decision loop is outside the machine. That distinction will matter later when we discuss autonomy and human control.

A robot is not one technology

Do not reduce robotics to programming. A correct program cannot make a disconnected motor move. A perfect gearbox cannot correct a bad sensor reading. A camera cannot save a frame that flexes enough to misalign the wheels.

Robotics sits where several systems collide:

• mechanics define what motion is physically possible;
• electrical systems deliver and control energy;
• software defines behavior;
• sensors produce evidence about the world;
• feedback changes behavior using that evidence;
• people define the mission, limits, and stop conditions.

That collision is the point of this course.

Your first artifact

Choose any robot you can observe from a photo, video, simulator, kit, household device, or public technical page. Draw a system map with at least these labels:

power → controller → actuator → physical motion

Add at least one sensor path if the robot has one. Then mark one human decision that remains outside the robot.

Do not worry about perfect symbols. The map should make the system easier to reason about than the original object did.