We consider a novel problem where multiple rigid convex polygonal objects
rest in randomly placed positions and orientations on a planar surface visible
from an overhead camera. The objective is to efficiently grasp and transport
all objects into a bin using multi-object push-grasps, where multiple objects
are pushed together to facilitate multi-object grasping. We provide necessary
conditions for frictionless multi-object push-grasps and apply these to filter
inadmissible grasps in a novel multi-object grasp planner. We find that our
planner is 19 times faster than a Mujoco simulator baseline. We also propose a
picking algorithm that uses both single- and multi-object grasps to pick
objects. In physical grasping experiments comparing performance with a
single-object picking baseline, we find that the frictionless multi-object
grasping system achieves 13.6\% higher grasp success and is 59.9\% faster, from
212 PPH to 340 PPH. See
https://sites.google.com/view/multi-object-grasping for videos and code.

本文研究了使用摄像头识别平面上的多个刚体凸多边形物体，利用多目标推动 - 抓取算法将它们可靠地移动到垃圾桶里。实验结果表明，使用摩擦力的多目标抓取系统比单目标抓取系统表现更好，速度大约快了 59.9％。

平面多目标抓取

Multi-Object Grasping in the Plane

Flexible pick-and-place is a fundamental yet challenging task within
robotics, in particular due to the need of an object model for a simple target
pose definition. In this work, the robot instead learns to pick-and-place
objects using planar manipulation according to a single, demonstrated goal
state. Our primary contribution lies within combining robot learning of
primitives, commonly estimated by fully-convolutional neural networks, with
one-shot imitation learning. Therefore, we define the place reward as a
contrastive loss between real-world measurements and a task-specific noise
distribution. Furthermore, we design our system to learn in a self-supervised
manner, enabling real-world experiments with up to 25000 pick-and-place
actions. Then, our robot is able to place trained objects with an average
placement error of 2.7 (0.2) mm and 2.6 (0.8){\deg}. As our approach does not
require an object model, the robot is able to generalize to unknown objects
while keeping a precision of 5.9 (1.1) mm and 4.1 (1.2){\deg}. We further show
a range of emerging behaviors: The robot naturally learns to select the correct
object in the presence of multiple object types, precisely inserts objects
within a peg game, picks screws out of dense clutter, and infers multiple
pick-and-place actions from a single goal state.

该研究提出一种应用基于机器学习的灵活取放技术实现对物体的自主操作方法，通过学习与仿真操作降低了对于物体模型的需求。