Computers Make Decisions Using Rules

Overview

So far, your robot has done exactly what you told it, in exactly the order you wrote it. Today that changes. You’ll write a program that makes a choice — it looks at a value and decides what to do based on a rule. The robot will sort freight: one action for one kind, a different action for another.

By the end of this activity you will be able to:

• Explain what a Condition is and how an if statement uses one to make a decision.
• Use if / else to make a robot take different actions for different values.
• Predict what a program will do by reading its rule before you run it.
• Connect rule-based decisions to how intelligent systems classify and respond to the world.

New This Time: Making a Decision

You already know variables, functions, and the commands k.motor(), k.msleep(), and k.ao(). Today you add one new tool: the if statement.

freight_type = 1             # a number YOU set

if freight_type == 1:        # IF freight_type is 1...
    drive_forward()           # ...do this
else:                         # OTHERWISE...
    turn_right()              # ...do this instead

freight_type = 1      # ONE equals  =   SETS the value to 1

if freight_type == 1:  # TWO equals  ==  ASKS "is it equal to 1?"

Phase 1 — Activate: The Bouncer at the Door

Imagine you are a bouncer at the door of an event. You have exactly one rule: if a person’s ticket says “VIP,” send them left; otherwise, send them right. You don’t know these people. You don’t judge them. You just check the ticket against the rule and act.

Send each person left or right by your rule:

A bouncer follows a rule without deciding who “deserves” what. How is that like the way a computer makes a decision?

Phase 2 — Concept: Conditions, Rules, and Classification

A Condition Is a True/False Question

Every decision a computer makes starts with a condition — a question that is either true or false. freight_type == 1 is a condition. Right now it is either true (the value really is 1) or false (it is anything else). There is no “maybe.” This is called Boolean thinking: every condition is true or false, nothing in between.

Classification

When a computer sorts things into groups using rules, that is classification. A mail machine reads a ZIP code and sends each letter to the right bin. Your program reads a freight value and sends the robot in the right direction. Same idea: a rule turns information into an action.

Pattern Recognition

Rules let a system respond to a pattern instead of a single fixed case. Your one rule — “if type 1, go straight; else turn” — handles every freight value you could ever set, not just one. Write the rule once, and it works for the whole pattern of cases.

In your own words: what is the difference between a program that follows a fixed list of steps and one that makes a decision?

Phase 3 — Plan

The Sorting Task

Step 1 — Write Your Rule in Plain English

Before any code, write your sorting rule as a sentence in the form “IF … THEN … OTHERWISE …”.

Step 2 — Predict the Two Outcomes

Fill in what the robot should do for each freight value.

Step 3 — Trace the Rule by Hand

Before you run anything, predict the robot’s action for each value. This is called tracing — following the rule the way the computer will.

Phase 4 — Build & Run

Starting Code Template

Type this program into your robot controller. The rule lives inside main(). Change freight_type at the top to test both paths. Fill in the actions to match your Phase 3 plan.

#!/usr/bin/python3

# Unit 1, Big Idea 3: Freight Sorter

# Name: _______________________   Date: ___________

import os, sys
sys.path.append("/usr/lib")
import _kipr as k

DRIVE_SPEED = 50

# A number YOU set. Think of it as the freight you're sorting.

# Change this value and the robot makes a different decision.
freight_type = 1     # try 1, then try 2

def drive_forward():
    k.motor(0, DRIVE_SPEED)
    k.motor(3, DRIVE_SPEED)
    k.msleep(1000)
    k.ao()

def turn_right():
    k.motor(0, DRIVE_SPEED)
    k.motor(3, -DRIVE_SPEED)
    k.msleep(600)
    k.ao()

def main():

    if freight_type == 1:       # IF the freight is type 1...
        drive_forward()          # ...send it straight ahead
    else:                        # OTHERWISE (it's not type 1)...
        turn_right()             # ...send it to the right instead

main()

Run It Both Ways — Results

Run the program with freight_type = 1, then change it to 2 and run again. Record what the robot actually did.

Checklist

• You used == (two equals) inside the if, not a single =
• Every BLOCK is indented to the same level
• You tested BOTH values, not just one
• The robot did something different for each value

Phase 5 — Debug

Decisions create a brand-new kind of bug: the robot does the wrong action, even though it runs without an error. That means the rule ran fine — but it was the wrong rule, or the value was not what you thought.

Debugging Log

Describe one decision bug you hit. Did the robot do the wrong thing, or refuse to run? How did you find the cause?

Phase 6 — Connect: The AI Literacy Bridge

Your robot did not “understand” the freight. It checked a value against a rule you wrote and acted. Almost every intelligent system works this way underneath: an email app checks features of a message against rules and decides “spam or not spam.” A photo app checks patterns and decides “face or not a face.” The system is not judging — it is classifying with rules. The intelligence is in the rules, and a human wrote them.

Read each scenario. Think it through, then write your answer.

A self-checkout machine decides whether the item you scanned matches its weight. What is the rule it is checking? What happens when the rule is wrong?

Your freight sorter only knows two outcomes: type 1, or everything else. A real sorting system might have ten kinds of freight. What is one problem with a rule that lumps everything that isn’t type 1 into a single “else”?

Who is responsible when an automated system classifies something wrong — sends the wrong freight, flags the wrong email? Connect this to the fact that a person wrote the rule.

Phase 7 — Individual Reflection

Complete this section on your own.

1. What is a condition? Write a definition in your own words.

2. Explain the difference between = and ==. Why does it matter inside an if statement?

3. Today your robot took different actions for different values without you rewriting the rule. Why is making a decision more powerful than a fixed list of steps?

4. Complete this in 2–3 sentences: “Intelligent systems identify patterns and use rules to make decisions. This means that when an AI system classifies something wrong…”

Extension Challenges

Finished early? Try one or more of these.

Extension A — A Third Choice

• Right now your rule has two outcomes. Add a middle case using elif: type 1 does one thing, type 2 does another, everything else does a third.
• Trace your new rule by hand for freight_type = 1, 2, and 9 before you run it.

Extension B — Flip the Rule

• Change your condition so the robot does the OPPOSITE — drives straight for everything except type 1.
• What did you change? Did you change the condition, the actions, or both?

Extension C — Greater Than

• == is not the only test. Try if (freight_type > 3) — “is the value greater than 3?”
• Predict, then test: what does the robot do for values 1, 3, and 7?

Extension D — Two Rules in a Row

• What happens if you write two separate if statements one after another, each checking freight_type?
• Could the robot ever do two actions in one run? When would that be useful, and when would it be a bug?

Extension E — Random Freight

• Real freight wouldn’t always be the same type every run. Add import random to the line after import _kipr as k, then use random.randint(1, 10) to generate a random freight_type each time your program runs.
• Run it several times. Does your sorting rule still make the correct choice across many random values?