CS 363: Homework


Current HW

HW 1: Your favorite robot

Due Tuesday, 22 September

Take a look at information about real robot research projects. Find one that you like and write a 1-page wiki entry on it, with links to the project and relevant papers. Identify and describe the mechanisms or joints, the sensors, computational abilities, and software capabilities. What about the robot makes it interesting?

A good place to start is the CMU Robotics Institute.

Please put the label cs363f09hw01 on your writeup.

HW 2: Bug algorithm

Due Tuesday, 29 September

Do a lit search and figure out what the family of Bug robot navigation algorithms are. Pick one of the Bug algorithms and write up a wiki page describing it along with links to resources/papers that describe it.

HW 3: Kalman filter exploration

Due Tuesday, 27 October

Download the files kalman1D.c and gaussDist.c. Put them in a directory for using the Nomad simulator. You can compile them using the command:

gcc -o kalman kalman1D.c gaussDist.c Nclient.c -lm

The program moves the robot to a specific distance from the wall using a sonar that has added noise.

  1. Read the code and figure out what the state variable is and how the Kalman filter works.
  2. Modify the code so it runs for a fixed number of iterations and writes out the true measurement, the noisy measurement, and the estimated state to a file on each iteration. Run the code and plot the three curves. As soon as the robot has stabilized its location relative to the wall, change the location of the wall once by moving it closer or further from the robot. Make sure you run for enough iterations for it to stabilize a second time.
  3. Using command line arguments you can change both the actual noise value (what is added to the signal) and the noise estimate for the measurement (R) used in the Kalman filter. See how making these different affects the response. Explore different ratios with the state noise level. Generate at least 3-4 more plots showing the different ways the robot responds.
  4. Put the plots up on a wiki page and describe how you think the response of the system changes with different actual and estimated measurement noise parameters.

Please put the label cs363f09hw03 on your writeup.