This article contains heuristics for following:
本文包含以下启发式方法:
- Parallel jaw grasps.
下颌平行抓紧。
- Suction grasps.
吸气。
- Linear push policies for improving parallel jaw grasps.
线性推动策略可改善平行颚的抓握力。
- Toppling policies for improving suction grasps.
改善吸力控制的主要政策。
Grasping is one of the fundamental subtask of a robotic manipulation pipeline. Both learning based and physics / geometry based grasping methods can benefit from grasp sampling heuristics in this article. Even if you are using a large arm farm to teach your robots the skills of grasping, you can save your robots quite a lot of time with these heuristics. This article summarizes the most common grasp sampling heuristics used in literature.
抓取是机器人操纵管线的基本子任务之一。 本文中的基于学习和基于物理/几何的抓取方法都可以从抓取采样启发式方法中受益。 即使您正在使用大型武装农场来教您的机器人掌握技巧 ,您也可以通过这些启发式方法为您的机器人节省大量时间。 本文总结了文献中最常见的抓取采样启发式方法。
Some of the common ways to use these heuristics are:
使用这些启发式方法的一些常见方法是:
-
Generating labels for learning based grasp planners (offline): 6-DOF GraspNet [4] uses these samplers for evaluation with physics based simulation. Grasps that retain the object between the gripper are considered successful after a predefined shaking motion. DexNet [2][3] evaluates these grasps based on analytic quasi-static grasp wrench space (GWS) analysis. Both methods score these sampled grasps based on how good they are in resisting disturbances. These scores are used as labels for training the grasp planners.
为学习型抓手计划者生成标签(离线): 6自由度GraspNet [4]使用这些采样器进行基于物理的仿真评估。 在预定义的摇动之后,将物体保持在抓具之间的抓握被认为是成功的。 DexNet [2] [3]基于准静态抓紧扳手空间(GWS)分析对这些抓握进行评估。 两种方法都基于它们在抵抗干扰方面的出色程度对这些采样的抓取力进行评分。 这些分数用作培训掌握计划者的标签。
-
During grasp synthesis (inference): DexNet [2][3] uses these sampled grasps as seeds for Cross Entropy Method (CEM), and optimizes grasps based on predicted grasp quality from GQ-CNN (Grasp Quality Convolutional Network). Traditional geometric methods, prune these candidate grasps if they are kinematically infeasible or if they result in collision between gripper and other objects or environment. The best of these samples are picked for execution.
在抓取合成(推断)期间: DexNet [2] [3]将这些采样的抓取用作交叉熵方法(CEM)的种子,并基于GQ-CNN(抓取质量卷积网络)的预测抓取质量来优化抓取。 传统的几何方法会修剪这些候选对象,如果它们在运动学上不可行,或者导致抓手与其他对象或环境之间发生碰撞。 从这些样本中最好的样本进行执行。
We will summarize the details of heuristics for each type of grippers used for manipulation.
我们将总结用于操纵的每种类型的抓取器的启发式方法的详细信息。
平行下颚抓 (Parallel jaw grasps)
Parallel jaw grasps jam the object between the grippers (Most often the grippers have rubber on them to increase the size of friction cones and thus the robustness of the grasp). Typically, the success of parallel jaw grasp depends on local geometry around the grasp point like if the grasp fits inside the gripper, friction btw gripper and object surface, mass of the object.
平行的下颌抓握器会夹住抓具之间的物体(大多数情况下,抓具上都带有橡胶以增加摩擦锥的大小,从而增加抓握的坚固性)。 通常,平行下颌抓取的成功取决于抓取点周围的局部几何形状,例如抓取是否适合抓取器内部,抓取器与物体表面的摩擦力,物体的质量。
Force Closure: If the contact points on the object are such that forces applied on those points don’t result in slippage and can resist gravity then force closure ( object doesn’t move with respect to the gripper ) is achieved, the grasp is considered successful.
力封闭:如果物体上的接触点不会对物体施加力而导致打滑,并且可以抵抗重力,则可以实现力封闭(物体相对于抓具不会移动),则应考虑抓紧成功。
Parametrization: Parallel Jaw Grasps are typically parametrized by 6-DOF pose of the gripper with initial configuration of open gripper.
参数化:平行下颌抓握通常是通过夹具的6自由度姿势和开放式夹具的初始配置来参数化的。
A Billion ways to grasp [1] summarizes several heuristics for parallel jaw grippers and evaluates their precision and coverage w.r.t a uniform sampler.
十亿种掌握方法[1]总结了平行颚式抓爪的几种启发式方法,并通过统一的采样器评估了它们的精度和覆盖率。
Assumption: Access to the 3D triangle mesh or 3D point cloud of the object so that surface normals can be computed.
假设:访问对象的3D三角形网格或3D点云,以便可以计算表面法线。
Here are the two most effective heuristics that are purely based on geometry:
以下是两个完全基于几何的最有效的启发式方法:
Approach based samplers:
基于方法的采样器:
These methods are characterized by approach vector of the gripper (red-dashed line) which typically aligns with normal to the palm (purple axis).
这些方法的特征在于抓手的接近向量(红色虚线),通常与手掌法线(紫色轴)对齐。
Pseudo code for approach based sampler:
基于方法的采样器的伪代码:
Antipodal based samplers:
基于对立采样器:
These methods sample directly on the space of possible contact points and try to exploit the grasps that create force closure.
这些方法直接在可能的接触点的空间上采样,并尝试利用产生力闭合的抓紧力。
Pseudo code for antipodal grasp sampler:
对偶采样器的伪代码:
Billion ways to grasp [1] evaluates grasps based on two metrics:
十亿种把握方式[1]根据两个指标评估把握:
-
Robust coverage: Percent of robust grasps (still successful in a small ϵ-neighborhood) sampled w.r.t oracle uniform sampler. This is very similar to recall.
稳健的覆盖率:稳健的掌握率(在一个小的ϵ邻域中仍然成功)从oracle统一采样器中采样的百分比。 这与召回非常相似。
-
Precision: Percent of the successful grasps among the sampled.
精度:样本中成功掌握的百分比。
As seen by the conclusion of Billion ways to grasp[1] from the tables, if you have a limited sampling budget antipodal sampling scheme provides both highest coverage and precision. However, asymptotically misses several ground truth grasps. These correspond to small scale features on objects and along the edges of objects.
从表中的“十亿种方法”的结论可以看出[1],如果您的采样预算有限,则对映采样方案可以提供最高的覆盖率和精度。 然而,渐进地错过了一些地面真理的掌握。 这些对应于对象上以及沿着对象边缘的小比例特征。
Visual illustration of what these sampled successful grasps and robust successful grasps look like. Each point is the grasp center and notice how robust grasps are clustered around object parts that fit nicely inside the gripper.
这些采样的成功掌握和稳健的成功掌握的视觉示意图看起来像。 每个点都是抓地力中心,注意坚固的抓地力如何聚集在非常适合抓手内部的对象部分周围。
吸盘 (Suction grippers)
Suction grippers form vacuum seal on the surface of the object and if that vacuum force is sufficient to resist the gravity and external wrenches, the grasp is robust. Typically suction grasp success depends on surface porousness, local geometry, mass and payload capacity of the suction gripper. These grippers are most popular for pick and place of objects in warehouse order fulfillment. DexNet 4.0 [6] which is one of the best published bin-picking system that uses composite policy between suction and parallel jaw grasps, chooses suction grasps for about 82% of attempts.
吸气夹具在物体表面形成真空密封,如果该真空力足以抵抗重力和外部扳手,则抓握力很强。 通常,抽吸抓取成功与否取决于抽吸抓具的表面Kong隙度,局部几何形状,质量和有效载荷容量。 这些抓取器最常用于在仓库订单履行中拾取和放置对象。 DexNet 4.0 [6]是使用率最高的垃圾收集系统之一,它在吸力和平行下颚抓地力之间采用了复合策略,大约有82%的尝试选择了吸力抓地力。
Parametrization: Suction grasps are typically parameterized by point p on the object surface and approach vector v as illustrated below.
参数化:吸气抓取通常由对象表面上的点p和逼近向量v参数化 ,如下所示。
Planarity Centroid Heuristic:
平面质心启发式:
Since successful suction grasps prefer planar non-porous surfaces, these heuristics try to find sufficiently planar surfaces on the object that are closer to COM (Center of Mass). Approach vectors are chosen along the surface normal because large motion tangential to surface might result in vacuum seal breakage.
由于成功的抽吸抓取更喜欢平面的无Kong表面,因此这些启发式方法试图在物体上找到更靠近COM(质心)的足够平面的表面。 沿表面法线选择接近向量,因为与表面相切的大运动可能会导致真空密封破裂。
Pseudo code for planarity centroid heuristic:
平面性质心启发式的伪代码:
Some examples of successful suction grasps on 3D meshes are visualized below.
下面是3D网格上成功吸取的一些示例。
DexNet 3.0 [3] evaluates suction grasps in physical robot trials based on two metrics:
DexNet 3.0 [3]根据两个指标评估物理机器人试验中的吸力抓地力:
-
Average Precision: Area under the precision / recall curve. How good is the heuristic in scoring high quality grasps ?
平均精度:精度/召回曲线下的面积。 高质量抓取得分的启发式方法有多好?
-
Success Rate: Fraction of the grasps that were successful.
成功率:成功把握的分数。
As can be seen from the table above, Planarity Centroid Heuristic does quite well compared to even learnt method DexNet 3.0 [3] on basic and typical objects.
从上表可以看出,相对于基本对象和典型对象的学习方法DexNet 3.0 [3],Planarity Centroid Heuristic的性能非常好。
Some of the failure cases of suction grasps are categorized as below:
吸力抓握的一些失败案例分类如下:
自适应采样器: (Adaptive sampler:)
These methods use heuristics that exploit the geometry to generate seed samples (described above) and further optimize the grasp according to grasp quality metric. Most often these are blackbox optimization technique such as cross entropy method (CEM) that doesn’t exploit object geometry. Although CEM is an optimization algorithm used in many areas, I would still consider it a heuristic since it doesn’t exploit the object geometries while sampling.
这些方法使用启发式技术,该启发式技术利用几何形状生成种子样本(如上所述),并根据抓地质量指标进一步优化抓地力。 最常见的是黑箱优化技术,例如不利用对象几何形状的交叉熵方法(CEM)。 尽管CEM是许多领域中使用的优化算法,但我仍然认为它是一种启发式方法,因为它在采样时不会利用对象的几何形状。
Additional assumption: Access to grasp quality function such as DexNet 2.0 / DexNet 3.0 Grasp Quality Network (GQ-CNN) or ability to evaluate quality of grasps in realtime based on GWS.
附加假设:可访问抓取质量功能(例如DexNet 2.0 / DexNet 3.0抓握质量网络(GQ-CNN))或基于GWS实时评估抓握质量的能力。
交叉熵法(CEM) (Cross Entropy Method (CEM))
If you were familiar with CEM, you may have noticed the use of GMM instead of Gaussians and this is because distribution of grasps on most objects are multi-modal.
如果您熟悉CEM,则可能已经注意到使用GMM而不是高斯模型,这是因为大多数对象上的控制点分布是多模式的。
Some examples of applying CEM method to DexNet 2.0 (parallel jaw grasps )and DexNet 3.0 (suction grasps) grasp quality functions to generate most robust grasps.
将CEM方法应用于DexNet 2.0(平行下颌抓紧)和DexNet 3.0(抽吸抓紧)的一些示例可抓握质量函数以生成最可靠的抓握。
增加把握的机会 (Improving chances of grasping)
Sometimes neither suction grasp not parallel jaw grasp is able to pick up any object in the heap. This is mostly due to inability to perceive robust grasps (occlusion) or inability to execute the perceived grasp ( collision or kinematic infeasibility ). In those cases non-prehensile ( fancy word for non-graspable ) actions are executed to either singulate the object to expose enough clearance for parallel jaw grasps or topple the object to expose a planar surface for suction grasps.
有时,吸力抓取器和平行颚抓持器都不能够拾取堆中的任何物体。 这主要是由于无法感知稳固的抓握(咬合)或无法执行感知的抓握(碰撞或运动学上的不可行性)。 在这些情况下,将执行非预紧(花哨词表示不可抓握)的操作,以将对象切成单个物体以露出足够的间隙以平行抓紧颚,或者翻倒对象以露出用于抓握的平面。
CAUTION: The following policies have not been tested on a real robot, so the results and conclusions don’t necessarily transfer.
注意:以下策略尚未在真实的机器人上进行过测试,因此结果和结论不一定会传递。
Parametrization: Push vector (p, q) where p = {x, y, z} starting point and q = {x’, y’, z’} is the end point.
参数化:推矢量(p,q) ,其中p = {x,y,z}的起点,而q = {x',y',z'}是终点。
线性推 (Linear Pushing)
Linear pushing policies typically help with separating the object heap so that parallel jaw grasps are accessible.
线性推入策略通常有助于分离对象堆,以便可以平行抓紧钳口。
Additional assumptions: Semantic instance segmentation of the objects on the bin so that each objects position on the bin is observed. Free space segmentation of the bin is also used in the linear pushing policies for choosing the push direction.
其他假设:容器上对象的语义实例分割,以便可以观察到每个对象在容器上的位置。 箱的自由空间分段还用于线性推动策略中,以选择推动方向。
Free Space Policy:
自由空间政策:
Aims to separate the two closest objects in the heap by pushing them towards the free space.
旨在通过将两个最接近的对象推向自由空间来分离它们。
Pseudo-code:
伪代码:
Boundary Shear Policy:
边界剪切策略:
Aims to separate two closest objects in the heap by pushing one of the objects along the boundary between the objects.
旨在通过沿对象之间的边界推动对象之一来分离堆中两个最接近的对象。
Pseudo-code:
伪代码:
Facilitating Grasping [5] evaluates above policies and few others in simulation in clearing the object heaps that don’t have accessible grasps and measures the confidence gain of both grasp types. As can be seen the linear pushing policies make the parallel jaw grasps more accessible than suction grasps.
促进抓取[5]在清除没有可访问的抓取的对象堆时评估了上述策略,并在仿真中评估了其他策略,并评估了两种抓取类型的置信度。 可以看出,线性推动策略使平行的下颚抓握比抽吸抓握更容易接近。
单一对象倒塌 (Singulated Object Toppling)
Facilitating grasping [5] also explores policies for toppling a singulated known 3D object so that quality of suction grasp after toppling can be improved.
促进抓握[5]还探讨了使单个已知3D对象发生倒塌的策略,从而可以提高倒塌后的吸取质量。
Assumptions: Known 3D object with known transition distribution of stable resting poses P[x_{t+1}|x_t, u_t] and access to suction grasp quality function V_s(x_t).
假设:已知的3D对象具有稳定的静止姿势P [x_ {t + 1} | x_t,u_t]的已知过渡分布,并可以访问吸力抓握质量函数V_s(x_t)。
Max Height Policy:
最大高度政策:
Highest possible point on the object that has surface normal within 15 degree of the supporting plane normal. This policy only gets executed if V_s(x_{t+1}) > V_s(x_t).
表面法线在支撑平面法线15度以内的物体上的最高点。 仅当V_s(x_ {t + 1})> V_s(x_t)时才执行此策略。
Greedy Policy:
贪婪政策:
Pick the action that makes the expected suction grasp more accessible.
选择使预期的吸力抓握更容易接近的动作。
Facilitating grasping [5] evaluates these policies in simulation and compares against a policy that runs complete value iteration based on P[x_{t+1}|x_t, u_t] and Vs(x_t).
便于掌握[5]在仿真中评估这些策略,并与基于P [x_ {t + 1} | x_t,u_t]和Vs(x_t)运行完整值迭代的策略进行比较。
Conclusion: This post explored different subtasks used for grasping and several effective heuristics for them. Please explore the references for more details on learning based / more effective policies. These heuristics are meant to provide intuition on each of the grasping subtasks and how they measure up to some of the more advanced methods.
结论:这篇文章探讨了用于掌握的不同子任务以及针对它们的几种有效启发式方法。 请浏览参考资料,以获取更多有关基于学习/更有效政策的详细信息。 这些试探法旨在为每个掌握的子任务提供直觉,以及它们如何衡量某些更高级的方法。
翻译自: https://medium.com/@darshanhegde_5567/heuristics-for-robotic-grasping-c28dbb90bce1
完美下巴标准