Research ArticleHUMAN-ROBOT INTERACTION

On the choice of grasp type and location when handing over an object

See allHide authors and affiliations

Science Robotics  13 Feb 2019:
Vol. 4, Issue 27, eaau9757
DOI: 10.1126/scirobotics.aau9757
  • Fig. 1 Taxonomy used to classify grasps in the experiment.

    The proposed taxonomy comprises three top-level categories: power, intermediate, and precision grasps. Power and precision grasps are both subdivided into circular and prismatic types. Further classifications are reported with a higher level of detail, leading to 15 classes (power prismatic: HW, PW, and M&L W; power circular: D and S; intermediate: L, ST, and W; precision prismatic: 4F, 3F, 2F, and 1F; precision circular: D, S, and T). Our analysis was focused on the abovementioned classes. For the sake of completeness, we also reported all 28 grasp types included in the classes and used to classify the grasps during the labeling process. For each grasp, the taxonomy reports a picture showing the grasp and an alphabetical label in reference to the taxonomy it was taken from: C from (3) and F from (8). The images are taken and adapted with permission from (8).

  • Fig. 2 Distribution of grasps throughout the experiment.

    The heat map on the left-hand side reports the occurrences of each grasp type over the 17 objects. For each object, 306 grasps were labeled. The histogram on the right-hand side shows the overall frequencies of the grasp types normalized by the total number of the labeled grasps (i.e., 5202). Precision grasps are the majority (62% of all grasps), followed by 30% of power grasps and 8% of intermediate grasps.

  • Fig. 3 Comparison of the performances of the passers in NIS and HS.

    (A) Histograms of the grasp choices over the two tasks; (B) histograms of grasp choices over the two conditions (passers in NIS and receivers in HS). (A′) and (B′) depict the same information but show higher levels of granularity in the grasp types. The frequencies shown are normalized by the total of 867 grasps performed by passers for each task of each session.

  • Fig. 4 Comparison of the performances of the passers in NIS and receivers in HS.

    (A) Histograms of the grasp choices over the two tasks; (B) histograms of grasp choices over the two conditions (passers in NIS and receivers in HS). (A′) and (B′) depict the same information but show higher levels of granularity in the grasp types. The frequencies shown are normalized by the total of 867 grasps performed by passers or receivers for each task of each session.

  • Fig. 5 Comparison of grasp types between passers and receivers during handovers.

    The heat map reports all 1734 combinations of the passer’s and receiver’s grasps observed in HS. The highest number of occurrences is found on the diagonal of this map. The red top-right semi-plane represents occurrences of more powerful grasps adopted by the receivers (with respect to passers) and is more populated than the bottom-left semi-plane (more powerful grasps by the passers with respect to receivers).

  • Fig. 6 Distribution of palm positions of the passers relative to the objects.

    Plots show the median approaching coordinate Pac of all the 17 passers relative to each object across sessions and tasks. The significant comparisons (P < 0.05) performed on |Pac| and on |dPac|, using the Wilcoxon test with the Bonferroni corrections, are reported with (*) and (★), respectively. In particular, |Pac| describes how the grasping locations are shifted toward the extremities of the objects, whereas |dPac| describes how they are clustered far from the median of the distribution. The bottom-right plot shows whether Key and Screwdriver were grasped by the handle or by the other extremity (Not Handle).

  • Fig. 7 Experimental setup and test objects set.

    (Top) Left: Outlines of the three types of objects that we used and their frame system identified by the axes x, y, and z along their major dimensions X, Y, and Z (XYZ), respectively. These drawings show, for each type of object, the distance vector Embedded Image from the centroids of the objects (CO) to the centroid of the passer’s hand (CH) and the approaching coordinate of the passer’s hand in the object frame (ac). Right: The experimental setup, the test objects, and the test gloves are shown. (Bottom) The table reports, for each object, its label, its three major dimensions, its mass, the mathematical definition of ac, and the tasks. We refer to dx, dy, and dz as the components along x, y, and z of the vector Embedded Image.

  • Table 1 Statistical results.

    Results of the comparisons performed using the Wilcoxon test on the absolute value of the passer’s palm position relative to the object (|Pac|) and the absolute distance between the median of the distribution of each condition and the passer’s palm (|dPac|). The P values reported are adjusted according to the Bonferroni correction. Significant results (P < 0.05) are boldface.

    NIS versus HSTask 1 versus task 2
    Task 1Task 2NISHS
    |Pac||dPac||Pac||dPac||Pac||dPac||Pac||dPac|
    CPenZ = −2.533
    P = 0.045
    Z = −0.544
    P > 0.9
    Z = −1.586
    P = 0.451
    Z = −2.580
    P = 0.040
    Z = −0.970
    P > 0.9
    Z = −1.728
    P = 0.336
    Z = −0.260
    P > 0.9
    Z = −0.724
    P > 0.9
    OPenZ = −2.959
    P = 0.012
    Z = −0.639
    P > 0.9
    Z = −3.432
    P = 0.002
    Z = −1.870
    P = 0.24
    Z = −1.160
    P > 0.9
    Z = −0.213
    P > P > 0.9
    Z = −0.781
    P > 0.9
    Z = −1.349
    P > 0.9
    KeyZ = −3.432
    P = 0.002
    Z = −1.396
    P > 0.9
    Z = −3.053
    P = 0.009
    Z = −0.876
    P > 0.9
    Z = −3.290
    P = 0.004
    Z = −2.012
    P = 0.177
    Z = −3.385
    P = 0.003
    Z = −0.260
    P > 0.9
    WShapeZ = −3.621
    P = 0.001
    Z = −2.722
    P = 0.026
    Z = −2.438
    P = 0.059
    Z = −0.355
    P > 0.9
    Z = Z = −3621
    P = 0.001
    Z = −2.689
    P = 0.029
    Z = −0.024
    P > 0.9
    Z = −0.260
    P > 0.9
    AppleZ = −2.296
    P = 0.087
    Z = −0.402
    P > 0.9
    Z = −2.012
    P = 0.177
    Z = −0.639
    P > 0.9
    Z = −0.686
    P > 0.9
    Z = −0.166
    P > 0.9
    Z = −0.544
    P > 0.9
    Z = −1.302
    P = 0.772
    BallZ = −1.160
    P > 0.9
    Z = −2.012
    P = 0.177
    Z = −2.296
    P = 0.087
    Z = −2.533
    P = 0.045
    Z = 0.213
    P > 0.9
    Z = −0.071
    P > 0.9
    Z = −3.148
    P = 0.007
    Z = −1.633
    P = 0.410
    HDiskZ = −3.337
    P = 0.003
    Z = −0.828
    P > 0.9
    Z = −1.349
    P = 0.709
    Z = −0.686
    P > 0.9
    Z = −3.527
    P = 0.002
    Z = −1.065
    P > 0.9
    Z = −0.639
    P > 0.9
    Z = −0.308
    P > 0.9
    LDiskZ = −3.624
    P = 0.001
    Z = −2.059
    P = 0.158
    Z = −0.828
    P > 0.9
    Z = −1396
    P = 0.651
    Z = −3.574
    P = 0.001
    Z = −1.965
    P = 0.199
    Z = −0.923
    P > 0.9
    Z = −1.870
    P = 0.246
    FGlassZ = −1.444
    P = 0.595
    Z = −1.775
    P = 0.303
    Z = −2.533
    P = 0.045
    Z = −3.14
    P = 0.007
    Z = −2.533
    P = 0.045
    Z = −2.391
    P = 0.067
    Z = −0.923
    P > 0.9
    Z = −0.592
    P > 0.9
    EGlassZ = −0.876
    P > 0.9
    Z = −0.639
    P > 0.9
    Z = −2.864
    P = 0.017
    Z = −2.864
    P = 0.017
    Z = −3.527
    P = 0.002
    Z = −2.485
    P = 0.052
    Z = −0.497
    P > 0.9
    Z = −0.970
    P > 0.9
    FBottleZ = −1.728
    P = 0.336
    Z = −2.296
    P = 0.088
    Z = −3.574
    P = 0.001
    Z = −2.959
    P = 0.012
    Z = −2.343
    P = 0.077
    Z = −1.396
    P = 0.652
    Z = −1.775
    P = 0.304
    Z = −1.349
    P = 0.708
    EBottleZ = −1.775
    P = 0.303
    Z = −2.627
    P = 0.034
    Z = −0.450
    P > 0.9
    Z = −0.355
    P > 0.9
    Z = −1.870
    P = −0.246
    Z = −2.249
    P = 0.098
    Z = −0.118
    P > 0.9
    Z = −0.308
    P > 0.9
    BoxZ = −1.349
    P = 0.709
    Z = −0.166
    P > 0.9
    Z = −1.018
    P = 1.235
    Z = −0.734
    P > 0.9
    Z = −3.053
    P = 0.009
    Z = −0.544
    P > 0.9
    Z = −2.296
    P = 0.0867
    Z = −0.308
    P > 0.9
    BookZ = 0.639
    P > 0.9
    Z = −2.343
    P = 0.076
    Z = −2.154
    P = 0.125
    Z = −1.207
    P > 0.9
    Z = −2675
    P = 0.030
    Z = −0.308
    P > 0.9
    Z = −1.491
    P = 0.543
    Z = −3.243
    P = 0.005
    ScrewdriverZ = −1.207
    P > 0.9
    Z = −3.574
    P = 0.001
    Z = −2.533
    P = 0.045
    Z = −2.769
    P = 0.023
    Z = −1870
    P = 0.246
    Z = −0.639
    P > 0.9
    Z = −1823
    P = 0.274
    Z = −0.828
    P > 0.9
    BarZ = −3.621
    P = 0.001
    Z = −0.718
    P > 0.9
    Z = −3.290
    P = 0.004
    Z = −3.290
    P = 0.004
    Z = −2.107
    P = 0.141
    Z = −3.290
    P = 0.004
    Z = 0.308
    P > 0.9
    Z = −0.118
    P > 0.9
    TeddyZ = −1.396
    P = 0.650
    Z = −1.586
    P = 0.451
    Z = −2.769
    P = 0.023
    Z = −1.065
    P > 0.9
    Z = −1.633
    P = 0.410
    Z = −0.024
    P > 0.9
    Z = −1.160
    P > 0.9
    Z = −1.633
    P = 0.410

Supplementary Materials

  • Supplementary Materials

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.mp4 format). Example of trials with the views of the object tracking.

    Files in this Data Supplement:

Stay Connected to Science Robotics


Editor's Blog

Navigate This Article