This is my Final Project assignment for the Udacity AI for Robotics course.

Here's a link to it working: https://youtu.be/afsXm6Om7ck This is obviously a work in progress...

I decided to try an extended kalman filter on it using a constant velocity/yaw rate model. I used the following tutorial to help me with this filter.
https://github.com/balzer82/Kalman Thank you for the great work, Dresden!

For catching up to the rogue bot, I just use the EKF motion portion to predict its next N positions and use the hunter's max speed to figure out how to intercept it.

If you are in this class, please don't just copy the code.

• you'll get more out of it by going through the struggle and...
• you can do a heck of a lot better than this.
• my first version used a PID with no kalman filtering at all and did just as well.

Note: The Bonus part of the Final Project was passed with the above problem using the following values: # Various motion noise for Q x_var = y_var = 1.5*dt # set for max speed theta_var = pi/8.*dt # Assuming max turn in a step v_var = 1.5 # set for max speed d_theta_var = .01 # assuming low acceleration

noise_est = 80. # Set extremely high for last rogue robot catch

Target bot 1 successfully caught in 270 measurements. Target bot 2 successfully caught in 898 measurements. Target bot 3 successfully caught in 669 measurements.