Knight Rover Robotics

When you want to learn about robotics where do you begin?

The Mind Project is launching its new “Knight Rover Robotics Initiative” with the goal to generate a robotics kit with low-cost, readily available hardware, a suite of innovative software, and instructional materials pitched at every level of expertise from novice to experienced programmer.

The heart of Knight Rover is the open source software “RObotics Orchestration Kit” (ROOK for short) created by Eric Thill, who has been the leader of The Mind Project robotics team since he was a sophomore at Illinois State in 2008.  ROOK is being created for use in educational settings and so that it will allow students to engage with the software regardless of their level of expertise. At the most basic level, students learn to use the software to make a robot perform certain behaviors. The deeper they dig the more they will learn about computer programming and the greater freedom they will achieve to re-write and expand the computational systems controlling their machine.

Like all of our physical robot projects, this one is designed to complement our “virtual” robotics labs and our other online curriculum in AI and robotics. How this robot project will integrate with specific online modules will be discussed below.


KR 1.0 (Knight Rover Version 1)

The ROOK software is remarkably flexible and can be used to control a wide range of different types of robotic hardware.  We have chosen to build our first Knight Rover robot (KR 1.0) using the GoPiGo hardware platform from Dexter Industries, which is relatively inexpensive and readily available.

KR 1.0 Software: ROOK
KR 1.0 Hardware: GoPiGo

The hardware kit uses basic sensors, motors, and other components along with a Raspberry Pi as its central processing unit (its “brain”). The team is working to add new hardware capabilities (sensors and end-affectors) and to develop code that will expand its "cognitive" powers.

The "KR 1.0 Development Page" will document the teams progress.

KR 1.0 Development Page

If you are a member of a robotics group, please let us know about your own projects.


The Mind Project’s primary focus is on the creation of interactive, online curriculum modules that introduce cognitive science – the interdisciplinary study of cognition (mental activity, both biological and mechanical) – to students with no previous background. Many teachers (and their students) lack either the time or the resources to work with physical robots (they are both expensive and time-consuming) and so we have created “virtual” robotics labs and online robotics instruction to help students learn about artificial intelligence and robotics online.

Working on a virtual lab for a couple hours is not the same as working with a physical robot, however. Our goal is to offer the best possible resources for teachers who have used our virtual labs and are ready to move on to the real thing.

Robots are not all designed in the same way, with the same “architecture.”  While today’s cutting edge AI and robotics systems use a wide range of different approaches to achieve machine intelligence, a good starting point for understanding robotic architecture is to get some hands-on experience working with two fundamentally different kinds of robot: “top-down” robots and “bottom-up” (or "behavior-based") robots.

We have two virtual robotics labs that introduce students to these two types of robotic design:

Virtual Robotics Lab ("Top Down")

Virtual Robotics Lab (Behavior-Based or "Bottom-Up")

One of the first goals for this team is to produce a robot built on the "behavior-based" architecture. Rodney Brooks of MIT was an earlier developer and advocate of this approach, when it was completely ignored by most prominant roboticists. This page has a video in which Brooks explains his inspiration:

"Pioneers of Behavior-Based Robotics"

The best introduction to behavior-based robots is to design one yourself! In only 20 minutes engaging with our virtual behavior-based lab you will design a number of different robots and begin to understand how a relatively small number of "stupid" simulus-response behaviors, when arranged in a hierarchy (what Brooks calls a "subsumption architecture") can produce remarkably intelligent behavior. This lab, created by Ken Livingston and Josh de Leeuw (at Vassar College), has a brief one-page introduction and a way to design-your own robot to try to replicate the behavior of the robot in the target video.