1 Earn Black Belt in Computer Programming!

The best way to learn with Karel is to go through the self-paced and self-graded Karel programming course in NCLab, using this textbook as a reference. Karel (shown in Fig. 1) is a determined robot who embarks on a quest to find his missing friend.

Fig. 1: Karel the Robot.

The course comes with many examples and tutorial videos that ensure that every student can progress at his or her own pace. Lesson plans and student journals are available. The instructor does not have to be an expert in computer programming by any means. The course has five parts and 175 engaging game levels that ensure an exciting and fulfilling learning experience for all students.

Recommended minimum age of students is 10 years because of keyboarding skills, and there is no upper age limit. In fact, many teachers use the Karel course to learn computational thinking skills themselves, and adult learners in general enjoy the course as much as K-12 students do.

A great complementary activity to solving game levels is creating your own mazes and games, and publishing them on the web for your friends to play. The lesson plans contain more details, and tutorial videos are available on NCLab’s home page.

1.1 White Belt

Every student begins his or her journey with solving seven levels in manual mode, earning the White Belt. These levels are designed to make the students familiar with the five basic commands Go, Right, Left, Get, and Put for the robot, with the four directions on the compass, with various game goals and limitations, as well as with the relative right and left (as seen from the point of view of the robot). Fig. 2 shows a sample level Goldmine from this section. Here, Karel’s task is to collect three gold nuggets and return to his home square, using at most 20 steps. Students guide the robot by clicking on buttons.

Fig. 2: Sample level Goldmine in the White Belt section.

1.2 Yellow Belt

The Yellow Belt has four Degrees and it requires the students to complete 63 levels in 9 sections. In Degree 1, students type elementary commands instead of clicking on buttons. Fig. 3 shows a sample level Phone where Karel needs to move the phone on the mark and enter his home square.

Fig. 3: Sample level Phone in Yellow Belt 1.

In Degree 2, students learn how to use the counting (repeat) loop. Fig. 4 shows a level Forgotten where Karel needs to pick up four books and put them in the bags.

Fig. 4: Sample level Forgotten in Yellow Belt 2.

In Degree 3, students learn about if-else conditions. Fig. 5 shows a sample level Storm where Karel needs to find his way home home by following a trace of coins and gears.

Fig. 5: Sample level Storm in Yellow Belt 3.

In the last Degree 4, students learn about the conditional (while) loop. Fig. 6 shows a sample level Tunnel where Karel must collect all keys and reach his home square.

Fig. 6: Sample level Tunnel in Yellow Belt 4.

1.3 Purple Belt

The Purple Belt has three Degrees with 70 game levels split into 10 sections. In Degree 1, students learn about custom commands. Fig. 7 shows a sample level Water where Karel must collect all water bottles in order to survive in the desert.

Fig. 7: Sample level Water in Purple Belt 1.

In Degree 2 students learn about variables. Fig. 8 shows a sample level Pipeline where Karel must count all missing pieces in the pipeline.

Fig. 8: Sample level Pipeline in Purple Belt 2.

In the last Degree 3 students learn about Python lists. Fig. 9 shows a sample level Rugs where Karel needs to move all rugs from one house to another, remembering their exact positions.

Fig. 9: Sample level Rugs in Purple Belt 3.

1.4 Black Belt

The Black Belt has 28 levels that are split into two Degrees. In Degree 1 students learn about recursion. Fig. 10 shows a sample level Bakery where Karel needs to eat all pies without using loops.

Fig. 10: Sample level Bakery in Black Belt 1.

Finally, in Black Belt 2 students apply the techniques and concepts they learned previously, to solve challenging problems. In Fig. 11 Karel needs to clean four oil blots using only 10 lines of code (this leads to three levels of nested loops).

Fig. 11: Sample level Hazard in Black Belt 2.

1.5 Extras and Bonuses

For extremely talented students, there is one more section with challenging problems where they are guaranteed to find their match. Fig. 12 shows a sample level Land Surveyor where Karel must calculate the area of a garden of an arbitrary shape, without entering it, just using his GPS coordinates.

Fig. 12: Sample level Land Surveyor in Extras and Bonuses.

2 Introduction

2.1 Objectives

Learn basic facts about the Karel language and its history.
Learn how Karel differs from other programming languages.
Learn what skills this course will give you.

2.2 Brief history

The educational programming language Karel the Robot was introduced by Richard E. Pattis in his book Karel The Robot: A Gentle Introduction to the Art of Programming in 1981. Pattis first used the language in his courses at Stanford University, and nowadays Karel is used at countless schools in the world. The language is named after Karel Čapek, a Czech writer who invented the word "robot" in his 1921 science fiction play R.U.R. (Rossum’s Universal Robots). Various implementations of the language that can be downloaded from the web are shown in Fig. 13.

Fig. 13: Various implementations.

The original Karel language was strongly influenced by Pascal, a popular language of the 1980s. Since Pascal is used much less today, we refreshed the Karel language and adjusted its syntax to be closer to Python, a premier programming language of modern engineering and science. Our changes made the Karel language much easier to use. For illustration, compare the original Karel program

BEGINNING-OF-PROGRAM

DEFINE turnright AS
BEGIN
    turnleft
    turnleft
    turnleft
END

BEGINNING-OF-EXECUTION
    ITERATE 3 TIMES
    BEGIN
        turnright
        move
    END
    turnoff
END-OF-EXECUTION

END-OF-PROGRAM

with its exact new Karel equivalent

  def turnright
      repeat 3
          left

  repeat 3
      turnright
      go

In fact, the new Karel has a built-in command right for the right turn, so the above program can be written using just three lines:

  repeat 3
      right
      go

The command right was not part of the original Karel language – it was added by us after a very careful consideration. The main reason for adding it was that it made Karel more pleasant to watch as he moves through more complex mazes. Without this command, every right turn took three left turns, and as a result, Karel resembled a raging tornado. Orthodox Karel fans can still define and use their own rightturn command.

We made a few additional changes to the language in order to make it more accessible to kids – beepers were replaced with gems, nuggets, spiders and other fun objects, Karel now has his home square in the maze, and there are new objects representing obstacles (which Karel must not run into or he explodes) and containers (where Karel can drop off collectible objects).

Longer commands were replaced with shorter ones, such as leftturn with left, move with go, pickbeeper with get, and putbeeper with put. Karel’s syntax is virtually identical to Python, with the exception of colons in conditions, loops, and new commands – they were left out since our youngest programmers had difficulty typing them.

2.3 Who is Karel?

Karel is a little robot who loves adventure. He crosses jungles and deserts, swims in the ocean, and climbs icy mountains. He can collect objects, put them into boxes, bags and other containers, and he tries not to hit walls and other obstacles. He was manufactured with only five simple commands:

go ... make one step forward.
get ... pick up an object that is beneath you.
left ... turn left.
right ... turn right.
put ... put an object on the ground.

He has many built-in sensors that allow him to check his immediate surroundings. Let’s start with general sensors:

north ... are you facing North?
home ... are you at home?
empty ... is your pocket empty?

Then there are sensors that detect various obstacles:

wall ... is wall ahead?
fire ... is fire ahead?
water ... is water ahead?

and there are many others. Next there are sensors that help the robot detect collectible objects beneath him:

gem ... is a gem beneath you?
nugget ... is a nugget beneath you?
apple ... is an apple beneath you?
ball ... is a ball beneath you?

And last there are containers where the robot can put collectible objects:

mark ... is a mark beneath you?
box ... is a box beneath you?
bag ... is a bag beneath you?

and many others. Check the Appendix for a complete list of all collectible objects, obstacles, and containers!

2.4 What will Karel teach you?

Computer programming skills are highly valued today, and they will be even more valued in the future. Karel is the perfect language for beginners. It will teach you how to design algorithms and write working computer programs without struggling with technical complications of mainstream programming languages. Karel will teach you all important concepts of modern procedural programming languages including correct code formatting, counting loops, if-else conditions, while loops, custom commands, functions, variables, lists, and recursion.

Thanks to Karel’s simplicity, you can be done with Karel 1 and Karel 2 fairly quickly. If you work hard, then in a month or two you will be ready to move on to the Python Turtle. The more levels you solve in Karel 1 and Karel 2, the easier your transition will be. In the Turtle course you will be drawing beautiful patterns using full Python syntax. And finally, when you have learned the Turtle, NCLab also has the advanced Python programming course for you.

2.5 Is Karel a toy language?

Absolutely not! Despite its playful appearance, Karel features all key concepts of modern procedural programming. Technically speaking, it is a complete Turing machine. As a matter of fact, the complexity of algorithms that you will encounter in this textbook ranges from very simple to very hard.

2.6 How does Karel differ from other programming languages?

The biggest conceptual difference between Karel and mainstream procedural programming languages such as Python, C, C++, Java or Fortran is that the robot does not know Math. This is because Math is not needed to understand how to design great algorithms and to translate them into efficient computer programs.

NCLab Community

Protected: Two chapters

Your Comment

TAGS

HOT TOPICS

LATEST COMMENTS

RELATED TOPICS