A reach & grasp approach for object manipulation with mechanical - - PowerPoint PPT Presentation

▶

Mar 25, 2023 265 likes •391 views

A reach & grasp approach for object manipulation with mechanical hands Daoud N., Touvet F., Gazeau J-P., Eskiizmirliler S., Zeghloul S., Maier M. A. Grasp Planning and Task Learning by Imitation Workshop IEEE IROS 2010, October 18, Taipei

SLIDE 1

A reach & grasp approach for object manipulation with mechanical hands

Daoud N., Touvet F., Gazeau J-P., Eskiizmirliler S., Zeghloul S., Maier M. A.

Grasp Planning and Task Learning by Imitation Workshop IEEE IROS 2010, October 18, Taipei

1 Tuesday, October 19, 2010

SLIDE 2

ABILIS : Bioinspired approaches for intelligent handling and gripping Naël Daoud Jean-Pierre Gazeau François Touvet Marc Maier

2 Tuesday, October 19, 2010

SLIDE 3

Materials

 16 dof 4 fingers fully actuated hand  Classical 6 dof industrial robot (Kuka)  4 types of objects

3 Tuesday, October 19, 2010

SLIDE 4

Methods

 The reach problem is solved by dedicated

LWPR Matching Units (MUs)

 The grasp configuration is selected

through all possibilities by a Genetic Algorithm

 The manipulation is considered as a serial

grasp configurations

4 Tuesday, October 19, 2010

SLIDE 5

Reach

 3d position given by a parametrized offset from the object

center of gravity

 3d orientation estimated by MUs  Final solution computed by the MU which learns Inverse

Kinematics of the Kuka manipulator using his direct kinematics.

pposition axis

Arbib et.al. (9) approach angle Paulignan et.al.(8)

5 Tuesday, October 19, 2010

SLIDE 6

Grasp quality criterion

 A triple criterion to evaluate grasp quality

T1: minimization of grasp forces
T2: maximization of manipulability

(Yoshikawa, T., 1994)

T3: maximization of distances to joints limits

6 Tuesday, October 19, 2010

SLIDE 7

Grasp strategy

 Define C, α, and hand

rientation

 Intersection between a

random line of the grasp plan containing C and the object surface gives P1

 P2 and P3 calculation is

straight forward

Grasp plane Grasp synthesis

7 Tuesday, October 19, 2010

SLIDE 8

Genetic algorithm parameters

 8 individuals is enough  Crossover and mutation probabilities

are fixed to 0.25

 Most of the time, convergence is

achieved in less than 50 iterations

 Optimal grasp is attained when the

algorithm gives 5 times the same individual

, ,

8 Tuesday, October 19, 2010

SLIDE 9

Simulation Results

Four objects grasps were tested with different result qualities

Object # of generations

Com. time [sec]

Cylinder 8-28 0.98-1.54 parallelepiped 14-69 1.23-4.03

Note: Results are influenced by object size, object type and α angle. with A1=0.467, A2=0.411, A3=0.475

9 Tuesday, October 19, 2010

SLIDE 10

Experimental Results

Equilateral grasp test by minimizing only grasp forces with A1=1, A2=0, A3=0 Reach & grasp of cylindrical object Grasp of a parallelepiped object

10 Tuesday, October 19, 2010

SLIDE 11

Conclusion

 A hybrid reach & grasp control approach combining bio-inspiration

(LWPR) and solution space exploration (by GA) was presented.

 Ongoing and future work focuses on three issues:

i) the concept of Matching Units has been expanded to the

multi-finger (up to 5 fingers) grasp problem (submitted to ICRA 2011). It provides object-dependent top & side grasp based on learned heuristic methods of grasp kinematics.

Behavioral grasp experimentations on healthy subjects are going

to be performed in the aim of extracting biomimetic grasp strategies to be learned by MUs.

ii) A new 4-finger hand with new-generation actuators will

replace the existing hand with the goal of controlling both movement kinematics and dynamics during object-dependent reach and grasp.

iii) The performance of hybrid, biomimetic and non- biomimetic

control schemes will be compared on the ABILIS platform.

11 Tuesday, October 19, 2010