MoveIt Task Constructor

../../_images/mtc_example.png

The Task Constructor framework provides a flexible and transparent way to define and plan actions that consist of multiple interdependent subtasks. It draws on the planning capabilities of MoveIt to solve individual subproblems in black-box planning stages. A common interface, based on MoveIt’s PlanningScene is used to pass solution hypotheses between stages. The framework enables the hierarchical organization of basic stages using containers, allowing for sequential as well as parallel compositions. For more details, please refer to the associated ICRA 2019 publication.

Getting Started

If you haven’t already done so, make sure you’ve completed the steps in Getting Started.

Installing MoveIt Task Constructor

Install From Source

Go into your catkin workspace and initialize wstool if necessary (assuming ~/ws_moveit as workspace path):

cd ~/ws_moveit/src
wstool init

Clone MoveIt Task Constructor and source dependencies:

git clone https://github.com/ros-planning/moveit_task_constructor.git
wstool merge moveit_task_constructor/moveit_task_constructor.rosinstall
wstool update

Install missing packages with rosdep:

rosdep install --from-paths . --ignore-src --rosdistro $ROS_DISTRO

Build the workspace:

catkin build

Running the Demo

The MoveIt Task Constructor package contains an example pick-and-place application. You can try it out by running:

roslaunch moveit_task_constructor_demo demo.launch

On the right side you should see the Motion Planning Tasks panel outlining the hierarchical stage structure of the example pick_place_task. When you select a particular stage, the list of successful and failed solutions will be shown in the right-most window. Selecting one of those solutions will start its visualization.

../../_images/mtc_show_stages.gif

Basic Concepts

The fundamental idea of MTC is that complex motion planning problems can be composed into a set of simpler subproblems. The top-level planning problem is specified as a Task while all subproblems are specified by Stages. Stages can be arranged in any arbitrary order and hierarchy only limited by the individual stages types. The order in which stages can be arranged is restricted by the direction in which results are passed. There are three possible stages relating to the result flow: generator, propagator, and connector stages:

Generators compute their results independent of their neighbor stages and pass it in both directions, backwards and forwards. An example is an IK sampler for geometric poses where approaching and departing motions (neighbor stages) depend on the solution.

Propagators receive the result of one neighbor stage, solve a subproblem and then propagate their result on to the neighbor on the opposite site. Depending on the implementation propagating stages can pass solutions forward, backward or in both directions separately. An example is a stage that computes a Cartesian path based on either a start or a goal state.

Connectors are don’t propagate any results, but rather attempt to bridge the gap between the resulting states of both neighbors. An example is the computation of a free-motion plan from one given state to another.

Additional to the order types, there are different hierarchy types allowing to encapsulate subordinate stages. Stages without subordinate stages are called primitive stages, higher-level stages are called container stages. There are three container types:

Wrappers encapsulate a single subordinate stage and modify or filter the results. An example is a filter stage that only accepts solutions of an IK solver that satisfy a certain constraint.

Serial Containers hold a sequence of subordinate stages and only consider end-to-end solutions as results. An example is a picking motion that consists of a sequence of coherent steps.

Parallel Containers combine set of subordinate stages and can be used for passing the best of alternative results, running fallback solvers or for merging multiple independent solutions. Examples are running alternative planners for a free-motion plan, picking objects with the right hand or with the left hand as a fallback, or moving the arm and opening the gripper at the same time.

../../_images/mtc_stage_types.png

Stages not only support solving motion planning problems. They can also be used for all kinds of state transitions, as for instance modifying the planning scene. Combined with the possibility of using class inheritance it’s possible to construct very complex behavior while only relying on a well-structured set of primitive stages.

Open Source Feedback

See something that needs improvement? Please open a pull request on this GitHub page