Mechanical and Nuclear Engineering

Segbot

The Segbot is a two wheeled, balancing robot.

This robot can turn, tilt, travel across the floor, and each of its wheels can turn independently of one another. We can measure the turn angle, the tilt angle, the wheel positions, and the coordinates of the Segbot location. The turn and tilt angles determine the Segbot’s orientation. Of course, the wheel positions and the turn angle are related. We need five pieces of information wheel angles, tilt, and location coordinates) to measure the Segbot’s orientation and position. There are only two motors or actuators, one on each wheel. Notice that there are five things we can vary and only two motors to do it. The Segbot is underactuated because there are more things it can do than available motors.

The Segbot is nonholonomic because the Segbot can only move in the direction its wheels can drive it. It cannot move sideways.

The dynamic equations relate the voltages sent to the motors to the Segbot location acceleration and the angular accelerations of the Segbot tilt and wheels. These dynamic equations are mathematically nonlinear which means that the Segbot is a nonlinear dynamic system.

Why we built the Segbot
  • It’s fun!
  • It provided an opportunity for a group of undergraduates and one graduate student to interact, plan, design, build, and test the device.
  • It introduced a 3D model into our lab.
  • We developed an application that used an Inertial Measurement Unit (IMU) for determining the orientation of the Segbot (turn and tilt).
  • We built a system that operated autonomously - which means that the on-board computer is responsible for everything. This autonomous operation requires WiFi and battery power.
  • The Segbot presents both linear and nonlinear control challenges.
  • We are interested in both stabilizing and tracking control.
  • We wanted to use an advanced but popular microcontroller as the “brains” for controlling the Segbot.
  • The Segbot originally was developed at the University of Illinois and the microcontroller consisted of the Texas Instruments (TI) C2000F28335 Delfino® floating-point microcontroller coupled with rate gyroscopes and accelerometers. We wanted to build a Segbot with a different microcontroller. http://coecsl.ece.illinois.edu/segbot/segbot.html
Resources

ACC presentation 2015 (pdf)
Drawings (pdf)
Parts list (excel)
VIs and myRIO personality
Solidworks 3D model