Picking Up the Pieces: Grasp Planning via Decomposition Trees - PowerPoint PPT Presentation

Loading...

PPT – Picking Up the Pieces: Grasp Planning via Decomposition Trees PowerPoint presentation | free to download - id: 2082ff-ZDc1Z



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Picking Up the Pieces: Grasp Planning via Decomposition Trees

Description:

Corey Goldfeder, Peter K. Allen, Claire Lackner, Raphael Pelosoff. Grasp Synthesis ... Must account for dynamics, soft contacts, non-fingertip contacts, ... – PowerPoint PPT presentation

Number of Views:43
Avg rating:3.0/5.0
Slides: 19
Provided by: sgol1
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Picking Up the Pieces: Grasp Planning via Decomposition Trees


1
Picking Up the Pieces Grasp Planning via
Decomposition Trees
  • Corey Goldfeder, Peter K. Allen, Claire Lackner,
    Raphael Pelosoff

2
Grasp Synthesis
  • High dimensional, nonlinear space
  • configuration space joints pose
  • grasp quality is not smooth
  • Difficult to model analytically
  • Must account for dynamics, soft contacts,
    non-fingertip contacts, material properties
  • Many constraints
  • Obstacles, hand kinematics and scale

3
Our approach
  • Simulation based grasp synthesis has many
    advantages
  • Space of all grasps is too large to explore fully
    in simulation
  • We want a subspace that contains many good grasps

4
GraspIt!
  • Grasp simulator for both robotic and human hands
  • Includes kinematics,dynamics
  • Real time 3D visualization
  • Efficiently computesgrasp quality

Graspit! A Versatile Simulator for Robotic
Grasping, IEEE Robotics and Automation Magazine,
11.4
5
Grasping By Parts
  • Automatic Grasp Planning Using Shape Primitives
    -Miller et. al.

6
Superquadrics
  • Simple volumetric primitive
  • Small parameter space (11 dimensions)
  • Preserves approximate normals

7
Split-Merge Decomposition
  • Segmentation and Superquadric Modeling of 3D
    Objects - Chevalier, Jaillet, Baskurt
  • We added nearest neighbor pruning to reduce
    complexity by a factor of n

8
Decomposition Trees
A model
8 levels of decomposition
the decomposition tree
9
Decomposition Trees
  • Building a tree from the bottom up
  • Pairwise merge of parts with least error

10
How Many Parts?
  • Use an error threshold?
  • Problem large superquadrics can swallow
    important features, like handles, without much
    error
  • Solution fixed number of parts
  • decompose all objects to n superquadrics
  • n is chosen experimentally for a given hand

11
Planner Overview
  • Decompose into tree with n leaves
  • Plan grasps on superquadrics
  • using entire tree, not just leaves
  • Simulate candidates on actual geometry, using
    GraspIt!
  • Rank results by grasp quality

12
Results
  • Planned multiple stable grasps for all our test
    objects

13
Results
  • Works even for objects difficult to represent
    with superquadrics

14
(No Transcript)
15
Difficulties
  • Assumes knowledge of object geometry
  • Superquadric decomposition is slow
  • Grasping a single part is done heuristically
  • Cannot plan candidates on parts from different
    branches of the tree

16
Do Trees Help?
  • Without a tree, some good grasps
  • With a tree, many good grasps
  • if a grasp is unsuitable, another good grasp can
    be substituted

17
Contributions
  • Fully automatic implementation of
    grasping-by-parts
  • Abstracts away fine features
  • Allows multiple parts to be planned on as a group

18
Future Work
  • Incorporate existing SVM planner for individual
    superquadrics
  • Speed up decomposition
  • Questions?
About PowerShow.com