Faculty ProjectsUpdated and Upgraded Hardware for Introductory Robotics Laboratory
Project Manager
Brenna Argall, Associate Professor of Computer Science; Associate Professor of Mechanical Engineering; and Associate Professor of Physical Medicine and Rehabilitation
Amount Requested
$8,000
Summary
The objective of this project is to update the hardware used in an undergraduate laboratory course that aims to expose early-stage students to the basic foundations for pioneering robotics, through a focus on artificial intelligence and machine learning. The hardware used in this course is now 12 years old, and its degradation due to time has reached the point of negatively impacting learning outcomes. This project will replace the robot hardware used in the course, removing pain points and adding new capabilities. Undergraduates interested in robotics, from a range of disciplines, are expected to benefit.
Planned Activities/Investments
In 2013, when I developed (with Murphy funding) and first taught the course that has since become CS/ME 301: Introductory Robotics Laboratory, the objectives were the following. (O1) To introduce into the Northwestern University curriculum a course that intersected robotics with Machine Learning (ML) and Artificial Intelligence (AI)—a fusion area essential for an innovative engineering curriculum, and at that time absent from the Northwestern curriculum. (O2) To introduce students to the practical challenges faced when working with real robots, and ignite passion for robotics research. (O3) To draw students from outside of and across engineering disciplines—including but not limited to Computer Science (CS), Electrical and Computer Engineering (ECE), Mechanical Engineering (ME), and Biomedical Engineering (BME). Since that time, the course has been taught a total of 10 times and has been assigned a permanent course number (cross-listed in both CS and ME), and since its third teaching has been at capacity at the quarter start each time it is taught.
Objective O2 is the focus of this proposal. The laboratory assignments currently require constructing the robots and programming their behaviors, and this hands-on approach provides students with a firsthand understanding of challenges in robotics, such as noisy sensor data and uncertain actuation, and the thrill in overcoming these challenges.
However, the hardware kits currently used in this laboratory course (the Bioloid Premium from Robotis) are in urgent need of replacement for several reasons:
- The hardware is 12 years old: it is outdated, and wearing out. For example, the microcontrollers are no longer stable, reducing the available sensors onboard from three to one, which limits functionality (and augments student frustration) and accordingly also limits the breadth of robot behaviors that might be developed.
- The robots rely on code run on a remote computer, and thus also a connection to that computer. Over the years we have tried multiple connection modalities (Zigbee radios, Bluetooth), with varying degrees of success. Interference between the multiple robot platforms running concurrently during class remains an issue, as does the distance over which the connection is stable, which impedes development.
- The dependence on desktop computers located in the Undergraduate Teaching Laboratories (classroom CG50) has proven a technological headache that sets back student work and furthermore places a cap on the number of student teams.
The objective of this funding request is to alleviate all of these challenges through the purchase of new robot hardware for the course (alleviating issue 1), that furthermore has an onboard computer (alleviating issues 2 and 3) and will allow the class size to expand. The new robot hardware also has manipulation capabilities (through a robotic arm and gripper, that attaches to the hexapod base), which will expand the educational component of the course (to include robot manipulation).
Alleviating these challenges and incorporating these expansions will further advance the learning objectives O1-O3 of the course.
Impact
This project will impact undergraduate students, from across multiple disciplines and with a range of backgrounds, who are interested in learning about robotics. With the new hardware made possible through this project, more class time will be devoted to robot behavior development and learning—because less time will be lost addressing malfunctioning hardware. The outcome will be a richer and more productive learning experience for our students.
In past teachings, the course has seen enrollment from students with prior robotics experience ranging from zero to extensive. The programming component of the course is intentionally accessible to those with a limited computer science background, to facilitate a draw of students from across engineering disciplines (pursuant of O3). (For example, in the first teaching of this course, a quarter of the students came from outside of engineering and nearly half were Freshmen or Sophomores.)
This course furthermore has a track record of fostering an interest in research. A total of 22 undergraduate students have transitioned to independent research conducted within my lab alone after taking this course.
We expect both of these trends to continue and amplify as a result of this project.
The impact of this project will be measured through (1) CTEC evaluation scores and anecdotal CTEC comments, including a comparison between recent prior teachings and future teachings of the course, (2) the distribution of student affiliations and their prior robotics experience, and (3) the number of students who contact the instructor about undergraduate research opportunities.
Sustainability
The course will continue to be taught, using hardware purchased via Murphy funding, for many years. (As did the original hardware, purchased via Murphy funding 12 years ago.) For minor upkeep costs such as replacement sensors and batteries, the Undergraduate Teaching Laboratories budget historically has been sufficient and willing to cover such costs.
Deliverables
The outcome of this project will be to integrate new hardware into the Introductory Robotics Course. This will be achieved by the following deliverables:
- Acquisition of the robot and computer hardware. (September 2024)
- Revamp the four laboratory assignments of the course to incorporate the new hardware capabilities. (October-November 2024)
- Deploy the new hardware and assignments in the Winter 2025 teaching of the course, and in all subsequent teachings of the course. (January 2025)
Budget Overview
Robot Hardware:
- Robot base (hexapod), with robotic arm, gripper, RaspberryPi 4B 8GB onboard computer, battery, charger, and sensors (one each of camera and ultrasonic included with the hexapod robot, plus two additional ultrasonic sensors).
- 15 HiWonder Hexapod Robots with Robotic Arm and Raspberry PI 4B 8GB: $789.99 (each)
- 30 HiWonder Glowy Ultrasonic Sensors: $7.99 (each)
Computer Accessories:
- Monitor, keyboard, and mouse to plug into the RaspberryPi computer onboard the robot, to facilitate direct (onboard) programming of the robot hardware.
- 15 ReHisk Portable Monitors with micro-HDMI: $199.99 (each)
- 15 Perixx Mini Keyboards: $39.99 (each)
- 15 Logitech Wired USB Mouses: $9.99 (each)
- Total per kit: $1,047.95 x 15 = $15,719.25
Total Budget Amount: $16,000
Matching Funds
The Departments of Computer Science and Mechanical Engineering jointly commit to providing 50 percent matching funds for this project. Department Chairs Samir Khuller (CS) and Wei Chen (ME) have approved this commitment.