Skip to content

@ThermalSoaring Thermal Soaring

Change the repository type filter

All

    Repositories list

    12 repositories

    • Report

      Public
      A repository to hold our report
      0•0•0•0•Updated Jun 9, 2016Jun 9, 2016

    • Poster

      Public
      The senior project poster required for Senior Seminar.
      0•0•0•0•Updated Jun 5, 2016Jun 5, 2016

    • Machine-Learning-Policy-Formation

      Public
      Dynamic programming / Q learning policy - table based and with neural interpolation.
      Python
      •0•0•0•0•Updated May 30, 2016May 30, 2016

    • ardupilot

      Public
      ArduPlane, ArduCopter, ArduRover source
      C++
      •
      GNU General Public License v3.0
      •18k•0•0•0•Updated May 30, 2016May 30, 2016

    • thermal-soaring

      Public
      The final thermal soaring code with all parts integrated
      Python
      •0•3•0•0•Updated May 20, 2016May 20, 2016

    • bayesian-learning

      Public
      From simple Bayesian demos to using GPR and Bayesian parameter estimation on live data from the Piccolo simulator
      Jupyter Notebook
      •0•1•0•0•Updated Apr 19, 2016Apr 19, 2016

    • autopilot-communication

      Public
      Repository for the relay/data crunching between Thermal Identification, Path Decision Making, and the Autopilot.
      Python
      •
      MIT License
      •0•1•0•0•Updated Apr 17, 2016Apr 17, 2016

    • RLDraft1a

      Public
      Using reinforcement learning to maximize energy gained from thermal.
      MATLAB
      •2•2•0•0•Updated Feb 16, 2016Feb 16, 2016

    • TD0RandomwalkDemo

      Public
      May actually be using TD(lambda). Outputs average RMS error of the values of the states in a random walk, using the TD reinforcement learning algorithm. Note that batch learning was NOT used, although it will be in future demos.
      MATLAB
      •0•0•0•0•Updated Dec 2, 2015Dec 2, 2015

    • RL-Multi-Armed-Bandit

      Public
      Compares effectiveness of epsilon greedy algorithms as the number of trials increases.
      MATLAB
      •0•1•0•0•Updated Nov 30, 2015Nov 30, 2015

    • RL-Policy-Iteration

      Public
      Learning agent finds its way through a maze to a reward square, using dynamic programming.
      MATLAB
      •0•1•0•0•Updated Nov 30, 2015Nov 30, 2015

    • Exploring-Starts-MC

      Public
      Learning agent learns how to play gambling game using Monte Carlo Simulation.
      MATLAB
      •0•0•0•0•Updated Nov 30, 2015Nov 30, 2015