マニピュレーターのモーションプランニング

RRT および剛体ツリーを使用したパス計画

マニピュレーターのモーションプランニングには、ロボットの自由度 (DOF) およびロボットモデルの運動学的拘束に基づいた高次元の空間でのパスの計画が含まれます。ロボットモデルの運動学的拘束は、rigidBodyTree オブジェクトとして指定されます。manipulatorRRT を使用して、Rapidly-Exploring Random Tree (RRT) アルゴリズムを使用したジョイント空間でパスを計画します。

オブジェクト

`manipulatorRRT`	Plan motion for rigid body tree using bidirectional RRT
`manipulatorStateSpace`	State space for rigid body tree robot models
`manipulatorCollisionBodyValidator`	Validate states for collision bodies of rigid body tree
`workspaceGoalRegion`	Define workspace region of end-effector goal poses

関数

すべて展開する

RRT

`plan`	Plan path using RRT for manipulators
`interpolate`	Interpolate states along path from RRT
`shorten`	Trim edges to shorten path from RRT

状態空間と状態バリデーター

`isStateValid`	Check if state is valid
`isMotionValid`	Check if path between states is valid

ワークスペースゴール領域

`sample`	Sample end-effector poses in world frame
`show`	Visualize workspace bounds, reference frame, and offset frame

トピック

Pick and Place Using RRT for Manipulators
Using manipulators to pick and place objects in an environment may require path planning algorithms like the rapidly-exploring random tree planner. The planner explores in the joint-configuration space and searches for a collision-free path between different robot configurations. This example shows how to use the manipulatorRRT object to tune the planner parameters and plan a path between two joint configurations based on a rigidBodyTree robot model of the Franka Emika™ Panda robot. After tuning the planner parameters, the robot manipulator plans a path to move a can from one place to another.
Pick-and-Place Workflow Using RRT Planner and Stateflow for MATLAB
This example shows how to setup an end-to-end pick-and-place workflow for a robotic manipulator like the KINOVA® Gen3.
Pick-and-Place Workflow in Gazebo Using Point-Cloud Processing and RRT Path Planning
Setup an end-to-end pick and place workflow for a robotic manipulator like the KINOVA® Gen3.
Plan Paths With End-Effector Constraints Using State Spaces For Manipulators
Plan a manipulator robot path using sampling-based planners like the rapidly-exploring random trees (RRT) algorithm.

注目の例

Pick and Place Using RRT for Manipulators

Using manipulators to pick and place objects in an environment may require path planning algorithms like the rapidly-exploring random tree planner. The planner explores in the joint-configuration space and searches for a collision-free path between different robot configurations. This example shows how to use the manipulatorRRT object to tune the planner parameters and plan a path between two joint configurations based on a rigidBodyTree robot model of the Franka Emika™ Panda robot. After tuning the planner parameters, the robot manipulator plans a path to move a can from one place to another.

ライブスクリプトを開く