Skip to content

Latest commit

 

History

History
96 lines (78 loc) · 2.42 KB

README.md

File metadata and controls

96 lines (78 loc) · 2.42 KB

Quick Start

Before running followings, you must load environment variables on each terminal using baxter.sh

Problem 1

Problem 1.A

After implementing the original DMPs on the problem_1.py, run it as follows:

roscd assignment_1/src/assignment_1
python3 problem_1.py -s a

Problem 1.B

After implementing the plan() function in dmp.py, run it as follows:

roscd assignment_1/src/assignment_1
python3 problem_1.py -s b

Problem 1.C

After implementing the learn() function in dmp.py, run it as follows:

roscd assignment_1/src/assignment_1
python3 problem_1.py -s c

Problem 1.D

After implementing the plan() function in dmp.py, run it as follows:

roscd assignment_1/src/assignment_1
python3 problem_1.py -s d

Problem 2

Problem 2.A

Train an original DMP with a 2-dimensional demonstration on the problem_2.py. You can run the code as follows:

roscd assignment_1/src/assignment_1
python3 problem_2.py -s a1

After implementing the improved DMPs on the problem_2.py, you can run the same problem as follows:

roscd assignment_1/src/assignment_1
python3 problem_2.py -s a2

Problem 2.B

After implementing the plan() function dmp.py for the improved DMP, run it as follows:

roscd assignment_1/src/assignment_1
python3 problem_2.py -s b

Problem 3

You first need to launch the Baxter robot:

roslaunch baxter_gazebo baxter_world.launch

You can test the robot by running the joint trajectory example program. Please, open another terminal session. After sourcing the environment variables as before, run following:

rosrun assignment_1 joint_trajectory_client.py -l right

Note that you need to specify the left or right arm using the '-l' tag.

Problem 3.A

Train the improved DMPs with joint-space demonstrations via problem_3.py:

roscd assignment_1/src/assignment_1
python3 problem_3.py -s a1

Reproduce it with the Baxter robot

roscd assignment_1/src/assignment_1
python3 problem_3.py -s a2

Problem 3.B

Train the improved DMPs with Cartesian-space demonstrations and reproduce it via problem_3.py:

roscd assignment_1/src/assignment_1
python3 problem_3.py -s b

Problem 3.C

Train the improved DMPs with Cartesian-space demonstrations, adapt the goal to another pose, and reproduce it via problem_3.py:

roscd assignment_1/src/assignment_1
python3 problem_3.py -s c