Robotics Task and Motion Planning Library
Project description
pytamp
Python Interface for the robot task and motion planning(TAMP) library
We will commit the code soon.
Features
- Pure python library
- Support only kinematic world, so not consider dynamics in this repo
- Support simple motion planning (RRT*, Cartesian planning)
- Render Robot mesh using matplotlib or trimesh.Scene
- Support 4 benchmarks for tamp
- Support Robot TAMP with MCTS [Ours]
Installation
Requirements
You need pygraphviz and python3-tk package to see a MCTS tree.
-
On Ubuntu 18.04 or 20.04, Download graphviz and graphviz-dev and python3-tk using
apt
sudo apt install graphviz graphviz-dev python3-tk
-
On Mac, Download graphviz using
brew
brew install graphviz
Install pytamp
pip install pytamp
SceneManager
-
Scene
You can manage entire scene using SceneManager class. We recommend using the SceneManager.
For example, You can manage robot, object, gripper pose or collision as well as visualize their geom.
In addition, You can compute cartesian or RRT-star motion planning.
You can see various examples inexamples/scene
directory.-
Render using trimesh.Scene
baxter sawyer iiwa14 Panda Doosan UR5e
-
-
Planning
You can see an animation of planning that visualizes trajectory in
examples/planning
directory.Cartesian (X1) RRT-star (X1) -
Benchmark
You can use the 4 benchmarks.
Benchmark1 Benchmark2 Benchmark3 Benchmark4
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.