Path/motion planning

CoppeliaSim offers path/motion planning functionality via a plugin wrapping the OMPL library. The plugin, courtesy of Federico Ferri, exports several API functions related to OMPL. Its source code can be found here.

Following points should be considered when preparing a path/motion planning task:

  • Decide of a start and goal state. When the path planning object is a serial manipulator, a goal pose (or end-effector position/orientation) is often provided instead of a goal state. In that case function simIK.getConfigForTipPose can be used to find one or several goal states that satisfy the provided goal pose.
  • Create a path planning task with simOMPL.createTask.
  • Select an algorithm with simOMPL.setAlgorithm.
  • Create the required state space, which can be composed as a compound object: simOMPL.createStateSpace and simOMPL.setStateSpace.
  • Specify which entities are not allowed to collide with simOMPL.setCollisionPairs.
  • Specify the start and goal state with simOMPL.setStartState and simOMPL.setGoalState.
  • Call simOMPL.setup once to initialize OMPL's data structures.
  • Compute one (or more) paths with simOMPL.compute; when called multiple times it will reuse previously computed data, if the chosen algorithm supports it, e.g.: multi-query planners such as PRM.
  • Destroy the path planning task with simOMPL.destroyTask.
  • Often, path planning is used in combination with inverse kinematics: in a pick-and-place task for instance, the final approach should usually be a straight-line path, which can be generated with simIK.generatePath.

    • Above procedure is the regular approach, which sometimes lacks flexibility. Additionally, following callback functions can be set-up:

  • simOMPL.setStateValidationCallback
  • simOMPL.setProjectionEvaluationCallback
  • simOMPL.setGoalCallback

    • Path optimality

    Most of the planners use randomized algorithms, which don't provide optimal (i.e. shortest) paths. Some of those are optimizing planners, meaning that they are still randomized, but will use the remaining planning time to improve the first solution found, thus giving an asymptotic optimality guarantee.

    If a short path is preferable, then an optimizing planner (such as RRT*, PRM*, ...) should be used.

    Multi-query planners

    When using multi-query planners (PRM, PRM*, ...), just call simOMPL.setStartState and simOMPL.compute again. Planning will re-use and expand the previously built roadmap.

     

    Make sure to refer to following demo scenes for additional details:

  • scenes/kinematics/8-computingJointAnglesForRandomPoses.ttt
  • scenes/3DoFHolonomicPathPlanning.ttt
  • scenes/6DoFHolonomicPathPlanning.ttt
  • motionPlanningDemo1
  • motionPlanningAndGraspingDemo.ttt