Skip to content

Implementation of the ICRA 2020 Locomanipulation Paper: "Finding Locomanipulation Plans Quickly in the Locomotion Constrained Manifold"

License

Notifications You must be signed in to change notification settings

stevenjj/icra2020locomanipulation

Repository files navigation

Avatar Locomanipulation

This package contains code for performing kinematics and dynamics computations to enable locomanipulation with the NASA Valkyrie robot.

Video Result

Click to view on youtube

System

Ubuntu 16.04 ROS Kinetic

Dependency Installation

Add the following projects to your catkin workspace:

cd catkin_ws/src
git clone https://github.com/ipab-slmc/eigenpy_catkin
git clone --branch v1.0.1-with-pinocchio-v2.1.3 https://github.com/stevenjj/hpp-fcl_catkin.git
git clone --branch v2.1.3-with-hpp-fcl-v1.0.1 https://github.com/stevenjj/pinocchio_catkin
git clone https://github.com/stevenjj/val_model
cd ~/catkin_ws
rosdep install --from-paths src --ignore-src -r -y
catkin build

Installation

To install this package

cd catkin_ws/src
git clone https://github.com/stevenjj/icra2020locomanipulation
catkin build

Run Example Scripts

To run example locomanipulation trajectories. Launch one of the following:

roslaunch avatar_locomanipulation door_open.launch
roslaunch avatar_locomanipulation cart_push.launch
roslaunch avatar_locomanipulation door_open_different_cost.launch
roslaunch avatar_locomanipulation cart_push_different_cost.launch
roslaunch avatar_locomanipulation door_open_neural_net.launch  # Needs the neural network weights
roslaunch avatar_locomanipulation cart_push_neural_net.launch  # Needs the neural network weights

To visualize the reachability and locomanipulability regions. Launch one of the following:

roslaunch avatar_locomanipulation viz_reachability_contact_transition_space.launch
roslaunch avatar_locomanipulation viz_reachability_end_effector_space.launch
roslaunch avatar_locomanipulation viz_locomanipulability_contact_transition_space.launch
roslaunch avatar_locomanipulation viz_locomanipulability_ee_space.launch

Get Neural Network Weights

Download the neural network weights from here and extract the contents to the nn_models folder.

Known Issue

Sometimes when running one of the example launch files, the planner immediately returns not finding a path. Relaunching the file again should work. I'm not sure of the issue as this bug appears to be intermittent on my system.

About

Implementation of the ICRA 2020 Locomanipulation Paper: "Finding Locomanipulation Plans Quickly in the Locomotion Constrained Manifold"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •