MTC功能包
Go to file
Robert Haschke 52fdf30433 simplify compute() API
- remove bool return value
- always create a solution trajectory, also in case of failures
- success/failure determined from solution.isFailure()

minor adjustments during cherry-pick
2018-05-30 21:01:34 +02:00
core simplify compute() API 2018-05-30 21:01:34 +02:00
msgs SolutionBase: rename 'name' to 'comment' 2018-05-30 20:25:42 +02:00
rviz_marker_tools rviz marker tools 2018-02-04 09:00:54 +01:00
visualization SolutionBase: rename 'name' to 'comment' 2018-05-30 20:25:42 +02:00
.gitignore add gitignore for vim swp 2017-02-28 11:12:18 +01:00
.prepare_overlay.sh travis fix 2018-04-15 11:24:44 +02:00
.travis.yml travis: enable ccache 2018-04-27 19:58:15 +02:00
LICENSE.txt add full license information 2017-12-12 21:24:32 +01:00
README.md provide a minimalistic README.md 2017-12-08 21:29:48 +01:00

moveit_task_constructor

An approach to forward multi-step manipulation planning

The framework is currently under development. The API is unstable and incomplete.

Feedback is very welcome.

This project enables the user to specify and plan complex manipulation actions in terms of successive planning stages.

Individual stages compute robot trajectories relative to their expected start or end. The resulting planning pipeline, i.e. Task, extends different candidate trajectories from key states (Generator stages) until it generated feasible trajectories that extend through all stages.

The current aim is to replace MoveIt's old pick&place pipeline and provide a transparent mechanism to enable and debug complex motion sequences.