Skip to content

Latest commit

 

History

History
126 lines (85 loc) · 3.42 KB

README.md

File metadata and controls

126 lines (85 loc) · 3.42 KB

reward-shaping-ttr

This repo focus on the research topic about reward shaping via optimal control theory. [since this repo only contains code file, the code may not run properly without certain large data file which is not uploaded. Remember to check .gitignore]

Basic Requirement

  1. Install miniconda3 (or anaconda if you like).
  2. Create a new conda environment and do all the following steps in this environment.
conda create --name [env_name] python=3.5
conda activate [env_name]
  1. Install missing python-packages, such as gym, numpy, tensorflow(or tensorflow-gpu), matplotlib, joblib, keras, mpi4py, cycler,
conda install [pkg_name]
pip install -U [pkg_name]
  1. Install Gazebo8, ROS kinetic, MATLAB2019a for you computer.
  2. Install gazebo-ros from gazebo website. Note the matching between gazebo version and ROS version. Here I use gazebo8 and ROS kinetic.

Trouble Shooting

  1. No module named rospy
    cd $CONDA_PREFIX
	mkdir -p ./etc/conda/activate.d
	mkdir -p ./etc/conda/deactivate.d
	touch ./etc/conda/activate.d/env_vars.sh
	touch ./etc/conda/deactivate.d/env_vars.sh

And edit env_vars.sh:

#!/bin/sh
export PYTHONPATH=$PYTHONPATH:/opt/ros/kinetic/lib/python2.7/dist-packages
  1. No module named rospkg
pip install -U rosdep rosinstall_generator wstool rosinstall six vcstools
  1. ImportError: dynamic module does not define module export function (PyInit__tf2)
  • First, create catkin_ws folder at whereever you want:
mkdir catkin_ws
cd catkin_ws
mkdir src
cd src

  • Second, download ROS package tf2, tf, hector_sensor_description from github into src folder

  • Third, re-build these packages using catkin build command:

catkin build -DPYTHON_EXECUTABLE=path_to_miniconda/envs/[env_name]/bin/python3.5
  • Finally, Edit $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh again:
source /opt/ros/kinetic/setup.bash
source path_to_catkin_ws/catkin_ws/devel/setup.bash
  1. sys/os/glnxa64/libstdc++.so.6: version `CXXABI_1.3.11' not found
cd  path_to_matlab/MATLAB/R2018b/sys/os/glnxa64
mkdir exclude
mv libstdc++.so.6* exclude/
ln -s /usr/lib/x86_64-linux-gnu/libstdc++.so.6 libstdc++.so.6
ln -s /usr/lib/x86_64-linux-gnu/libstdc++.so.6.0.xx libstdc++.so.6.0.xx
  1. If you meet problems with xacro, saying no module named "glob" or "re", you need download xacro from github into catkin_ws src folder and compile it using python3.5 manully.
  2. Everytime you build something new in catkin_ws, remember to source the env_vars again.
  3. For this project, you also need edit on env_vars.sh like:
export PROJ_HOME=/local-scratch/xlv/reward_shaping_ttr
export QUADROTOR_WORLD_AIR_SPACE=$PROJ_HOME/worlds/air_space.world
export DUBINS_CAR_WORLD_CIRCUIT_GROUND=$PROJ_HOME/worlds/circuit_ground.world

export ROS_HOSTNAME=localhost
export ROS_MASTER_URI=http://localhost:11311

alias killgazebogym='killall -9 rosout roslaunch rosmaster gzserver nodelet robot_state_publisher gzclient'

Additional tips

  1. After you clone this repo to your local, here is a few things you may need do
  • change the $PROJ_HOME in env_vars.sh
$PROJ_HOME = 'path you place this repo at'
  • you may need to recompile catkin_ws folder to let it recognize your filepath
  • everytime you build something new in catkin_ws or modify something in env_vars, remember
source env_vars.sh
  1. stay tuned...