How Robots Think

Science Interactive lesson Free to play

How robots think is best understood as a repeating loop of three steps: Sense, Think, Act. A robot uses cameras and sensors to gather information about its surroundings, processes that information against its programmed instructions, and then moves or acts on the result — before starting the loop again. A robot's 'thinking' is not like a human's: it is an algorithm, a step-by-step list of instructions written by a programmer, much like a recipe the robot follows exactly.

Understanding this matters because robots are now part of everyday life in Singapore — cleaning robots at hawker centres, delivery robots on footpaths, and robot vacuums at home. Knowing that these machines follow instructions rather than make free choices helps children reason about technology instead of treating it as magic.

The key ideas to grasp are the Sense–Think–Act loop, algorithms as precise instruction lists, IF–THEN rules that let robots respond to situations (IF a wall is detected, THEN turn), and machine learning, where modern robots find patterns in many examples — but only know what their examples have shown them.

▶ Play the lesson — free, no signup

Want to create your own Spark? Sign up free — type any skill and LearnBuddy builds you a playable lesson.

Sign up free to create your own Spark

What this Spark covers

Frequently asked questions

Do robots actually think like humans?
No. A robot's 'thinking' is an algorithm — a step-by-step list of instructions written by programmers. Robots do exactly what their program says, nothing more and nothing less, so what looks like a decision is really an IF–THEN rule being followed.
What is the Sense–Think–Act loop?
It is the basic cycle every robot runs: sense the world through cameras and sensors, process that input against its instructions, then act by moving or doing something. The loop repeats continuously, which is how a robot vacuum keeps avoiding walls as it cleans.
What is an algorithm, in simple terms a child can understand?
An algorithm is like a recipe: a precise, ordered list of steps that must be followed exactly. Computer scientists write algorithms to tell robots and computers what to do, and if a step is wrong or missing, the robot gets it wrong too.
How do robots make choices, like turning when they hit a wall?
Programmers give robots IF–THEN rules: IF the sensor detects a wall, THEN turn and continue. The robot is not really choosing — it is matching the situation it senses to a rule it was given in advance.
What is machine learning, and can robots learn anything?
Machine learning lets some modern robots find patterns from many examples instead of following only hand-written rules. But they can only learn what their examples show them — a robot trained only on pictures of cats will not recognise a dog.

More Sparks like this

Related practice papers

Related reads

Loved this Spark? Sign up free for AskBuddy AI tutoring, past-year papers, and unlimited Sparks.

Sign up free →