Humanlike physics understanding for autonomous robots
自主机器人的类人物理理解
基本信息
- 批准号:EP/R031193/1
- 负责人:
- 金额:$ 38.62万
- 依托单位:
- 依托单位国家:英国
- 项目类别:Research Grant
- 财政年份:2018
- 资助国家:英国
- 起止时间:2018 至 无数据
- 项目状态:已结题
- 来源:
- 关键词:
项目摘要
How do you grasp a bottle of milk, nestling behind some yoghurt pots, within a cluttered fridge? Whilst humans are able to use visual information to plan and select such skilled actions with external objects with great ease and rapidity - a facility acquired in the history of the species and as a child develops - *robots struggle*. Indeed, whilst artificial intelligence has made great leaps in beating the best of humanity in tasks such as chess and Go, the planning and execution abilities of today's robotic technology is trumped by the average toddler. Given the complex and unpredictable world within which we find ourselves situated, these apparently trivial tasks are the product of highly sophisticated neural computations that generalise and adapt to changing situations: continually engaging in a process of selecting between multiple goals and action options. Our aim is to investigate how such computations could be transferred to robots to enable them to manipulate objects more efficiently, in a more human-like way than is presently the case, and to be able to perform manipulation presently beyond the state of the art.Let us return to the fridge example: You need to first decide what yoghurt pot is best to remove to allow access to the milk bottle and then generate the appropriate movements to grasp the pot safely- the *pre-contact *phase of prehension. You then need to decide what type of forces to apply to the pot (push it to the left or the right, nudge it or possibly lift it up and place the pot on another shelf etc) i.e. the *contact* phase. Whilst these steps happen with speed and automaticity in real time, we will probe these processes in laboratory controlled situations to systematically examine the pre-contact and contact phases of prehension to determine what factors (spatial position, size of pot, texture of pot etc) bias humans to choose one action (or series of actions) over other possibilities. We hypothesise that we can extract a set of high level rules, expressed using qualitative spatio-temporal formalisms which can capture the essence of such expertise, in combination with more quantitative lower-level representations and reasoning. We will develop a computational model to provide a formal foundation for testing hypotheses about the factors biasing behaviour and ultimately use this model to predict the behaviour that will most probably occur in response to a given perceptual (visual) input in this context. We reason that a computational understanding of how humans perform these actions can bridge the robot-human skill gap. State-of-the-art robot motion/manipulation planners use probabilistic methods (random sampling e.g. RRTs, PRMs, is the dominant motion planning approach in the field today). Hence, planners are not able to explain their decisions, similar to the "black box" machine learning methods mentioned in the call which produce inscrutable models. However, if robots can generate human-like interactions with the world, and if they can use knowledge of human action selection for planning, then this would allow robots to explain why they perform manipulations in a particular way, and also facilitate "legible manipulation" - i.e. action which is predictable by humans since it closely corresponds to how humans would behave, a goal of some recent research in the robotics community. The work will shed light on the use of perceptual information in the control of action - a topic of great academic interest and simultaneously have direct relevance to a number of practical problems facing roboticists seeking to control robots working in cluttered environments: from a robot picking items in a warehouse, to novel surgical technologies requiring discrimination between healthy and cancerous tissue.
在一个杂乱的冰箱里,你如何抓住一瓶躺在酸奶罐后面的牛奶?虽然人类能够使用视觉信息来计划和选择这种熟练的动作与外部对象非常容易和快速-一个设施获得的历史物种和作为一个孩子的发展- * 机器人斗争 *。事实上,虽然人工智能在国际象棋和围棋等任务中击败了人类的最佳表现,但当今机器人技术的规划和执行能力却被一般的蹒跚学步的孩子所超越。考虑到我们所处的复杂和不可预测的世界,这些看似微不足道的任务是高度复杂的神经计算的产物,这些神经计算概括并适应不断变化的情况:不断参与在多个目标和行动选项之间进行选择的过程。我们的目标是研究如何将这种计算转移到机器人身上,使它们能够以比目前更像人类的方式更有效地操纵物体,并且能够执行目前超出最先进水平的操纵。让我们回到冰箱的例子:你需要首先决定什么酸奶罐是最好的删除,以允许访问奶瓶,然后产生适当的运动,以把握锅安全-* 预接触 * 阶段的扩张。然后,你需要决定什么类型的力量施加到锅(推它到左边或右边,轻推它或可能举起它,并把锅放在另一个架子上等)即 * 接触 * 阶段。虽然这些步骤在真实的时间内以速度和自动性发生,但我们将在实验室控制的情况下探索这些过程,以系统地检查接触前和接触阶段的接触,以确定哪些因素(空间位置,锅的大小,锅的质地等)使人类选择一个动作(或一系列动作)而不是其他可能性。我们假设,我们可以提取一组高层次的规则,表示使用定性的时空形式主义,可以捕捉这种专业知识的本质,结合更定量的低层次的表示和推理。我们将开发一个计算模型,提供一个正式的基础测试的因素偏置行为的假设,并最终使用该模型来预测的行为,将最有可能发生在响应给定的感知(视觉)输入在这种情况下。我们的理由是,对人类如何执行这些动作的计算理解可以弥合机器人与人类的技能差距。最先进的机器人运动/操纵规划器使用概率方法(随机采样,例如RRT、PRM,是当今该领域的主要运动规划方法)。因此,规划者无法解释他们的决定,类似于电话中提到的“黑匣子”机器学习方法,它产生了不可理解的模型。然而,如果机器人可以与世界产生类似人类的交互,并且如果它们可以使用人类行为选择的知识进行规划,那么这将允许机器人解释为什么它们以特定的方式执行操作,并且还有助于“清晰的操作”-即人类可预测的行为,因为它与人类的行为密切相关,这是机器人领域最近研究的目标。这项工作将阐明在行动控制中使用感知信息-这是一个具有极大学术兴趣的话题,同时与机器人学家寻求控制在混乱环境中工作的机器人所面临的许多实际问题直接相关:从机器人在仓库中挑选物品,到需要区分健康和癌组织的新型手术技术。
项目成果
期刊论文数量(9)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Frontal theta brain activity varies as a function of surgical experience and task error.
- DOI:10.1136/bmjsit-2020-000040
- 发表时间:2020
- 期刊:
- 影响因子:0
- 作者:Balkhoyor AM;Awais M;Biyani S;Schaefer A;Craddock M;Jones O;Manogue M;Mon-Williams MA;Mushtaq F
- 通讯作者:Mushtaq F
Combining Coarse and Fine Physics for Manipulation using Parallel-in-Time Integration
使用并行时间积分将粗略和精细物理相结合进行操作
- DOI:10.48550/arxiv.1903.08470
- 发表时间:2019
- 期刊:
- 影响因子:0
- 作者:Agboh Wisdom C.
- 通讯作者:Agboh Wisdom C.
Pushing Fast and Slow: Task-Adaptive Planning for Non-prehensile Manipulation Under Uncertainty
快推和慢推:不确定性下非综合操纵的任务自适应规划
- DOI:
- 发表时间:2018
- 期刊:
- 影响因子:0
- 作者:Agboh W
- 通讯作者:Agboh W
Algorithmic Foundations of Robotics XIII - Proceedings of the 13th Workshop on the Algorithmic Foundations of Robotics
机器人算法基础 XIII - 第 13 届机器人算法基础研讨会论文集
- DOI:10.1007/978-3-030-44051-0_10
- 发表时间:2020
- 期刊:
- 影响因子:0
- 作者:Agboh W
- 通讯作者:Agboh W
Parareal with a learned coarse model for robotic manipulation
Parareal 具有用于机器人操作的学习粗略模型
- DOI:10.1007/s00791-020-00327-0
- 发表时间:2020
- 期刊:
- 影响因子:0
- 作者:Agboh W
- 通讯作者:Agboh W
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Anthony Cohn其他文献
Cognitive Workflow Capturing and Rendering with On-Body Sensor Networks (COGNITO)
使用体上传感器网络 (COGNITO) 进行认知工作流程捕获和渲染
- DOI:
- 发表时间:
- 期刊:
- 影响因子:0
- 作者:
Gabriele Bleser;Luis Almeida;Ardhendu Behera;Andrew Calway;Anthony Cohn;D. Damen;Hugo Domingues;Andrew Gee;Dominic Gorecky;David Hogg;Michael Kraly;Trivisio Prototyping;GmbH;Germany Gustavo;Maçães;Frédéric Marin;Walterio W. Mayol;M. Miezal;K. Mura;Nils Petersen;N. Vignais;Luís Paulo;Santos;G. Spaas;Germany Gmbh;Stricker - 通讯作者:
Stricker
Research Challenges and Opportunities in Knowledge Representation
知识表示的研究挑战和机遇
- DOI:
- 发表时间:
2013 - 期刊:
- 影响因子:0
- 作者:
Natasha Noy;Deborah L. McGuinness;Eyal Amir;Chitta Baral;Michael Beetz;S. Bechhofer;C. Boutilier;Anthony Cohn;J. Kleer;Michel Dumontier;Tim Finin;Kenneth D. Forbus;Lise Getoor;Yolanda Gil;J. Heflin;P. Hitzler;Craig A. Knoblock;Henry Kautz;Yuliya Lierler;Vladimir Lifschitz;Peter F. Patel;C. Piatko;D. Riecken;M. Schildhauer - 通讯作者:
M. Schildhauer
More needles less pain: The use of local anaesthesia during emergency arterial sampling
- DOI:
10.1016/j.joad.2016.03.013 - 发表时间:
2016-05-01 - 期刊:
- 影响因子:
- 作者:
Ruslan Zinchenko;Nicolaas Jacobus Prinsloo;Anton Zarafov;Maciej Grzesiak;Anthony Cohn - 通讯作者:
Anthony Cohn
Anthony Cohn的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Anthony Cohn', 18)}}的其他基金
The Detection of Archaeological residues using Remote Sensing Techniques (DART)
使用遥感技术 (DART) 检测考古残留物
- 批准号:
AH/H032673/1 - 财政年份:2010
- 资助金额:
$ 38.62万 - 项目类别:
Research Grant
MAPPING THE UNDERWORLD: MULTI-SENSOR DEVICE CREATION, ASSESSMENT, PROTOCOLS
绘制地下世界:多传感器设备创建、评估、协议
- 批准号:
EP/F06585X/1 - 财政年份:2009
- 资助金额:
$ 38.62万 - 项目类别:
Research Grant
Geometric Abstractions for Scalable Program Analyzers
可扩展程序分析器的几何抽象
- 批准号:
EP/G025177/1 - 财政年份:2008
- 资助金额:
$ 38.62万 - 项目类别:
Research Grant
相似国自然基金
Understanding complicated gravitational physics by simple two-shell systems
- 批准号:12005059
- 批准年份:2020
- 资助金额:24.0 万元
- 项目类别:青年科学基金项目
Chinese Physics B
- 批准号:11224806
- 批准年份:2012
- 资助金额:24.0 万元
- 项目类别:专项基金项目
Science China-Physics, Mechanics & Astronomy
- 批准号:11224804
- 批准年份:2012
- 资助金额:24.0 万元
- 项目类别:专项基金项目
Frontiers of Physics 出版资助
- 批准号:11224805
- 批准年份:2012
- 资助金额:20.0 万元
- 项目类别:专项基金项目
tau轻子衰变与新物理模型唯象研究
- 批准号:11005033
- 批准年份:2010
- 资助金额:18.0 万元
- 项目类别:青年科学基金项目
Chinese physics B
- 批准号:11024806
- 批准年份:2010
- 资助金额:24.0 万元
- 项目类别:专项基金项目
新强子态研究
- 批准号:10905053
- 批准年份:2009
- 资助金额:18.0 万元
- 项目类别:青年科学基金项目
高能强子对撞机Higgs衰变到双光子末态的寻找
- 批准号:10975134
- 批准年份:2009
- 资助金额:40.0 万元
- 项目类别:面上项目
强子对撞机上新物理信号的多轻子末态研究
- 批准号:10675110
- 批准年份:2006
- 资助金额:36.0 万元
- 项目类别:面上项目
强子对撞物理中的R宇称现象学研究
- 批准号:10575095
- 批准年份:2005
- 资助金额:28.0 万元
- 项目类别:面上项目
相似海外基金
RII Track-4:NSF: Physics-Informed Machine Learning with Organ-on-a-Chip Data for an In-Depth Understanding of Disease Progression and Drug Delivery Dynamics
RII Track-4:NSF:利用器官芯片数据进行物理信息机器学习,深入了解疾病进展和药物输送动力学
- 批准号:
2327473 - 财政年份:2024
- 资助金额:
$ 38.62万 - 项目类别:
Standard Grant
CAREER: Physics-Informed Deep Learning for Understanding Earthquake Slip Complexity
职业:基于物理的深度学习用于理解地震滑动的复杂性
- 批准号:
2339996 - 财政年份:2024
- 资助金额:
$ 38.62万 - 项目类别:
Continuing Grant
Collaborative Research: WoU-MMA: Understanding the Physics and Electromagnetic Counterparts of Neutrino Blazars with Numerical Simulations
合作研究:WoU-MMA:通过数值模拟了解中微子耀变体的物理和电磁对应物
- 批准号:
2308090 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
Standard Grant
Collaborative Research: NSFGEO-NERC: Understanding surface-to-bed meltwater pathways across the Greenland Ice Sheet using machine-learning and physics-based models
合作研究:NSFGEO-NERC:使用机器学习和基于物理的模型了解格陵兰冰盖的地表到床层融水路径
- 批准号:
2235052 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
Standard Grant
Understanding the Heterogeneity of Nanoscale Extracellular Vesicles, Exomeres, and Supermeres using Next Generation Optical Nanotweezers
使用下一代光学纳米镊子了解纳米级细胞外囊泡、外泌体和 Supermeres 的异质性
- 批准号:
10714221 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
CAREER: Understanding the Physics of Turbulent Flow, Erosion and Depositional Patterns in River Systems
职业:了解河流系统中湍流、侵蚀和沉积模式的物理原理
- 批准号:
2239550 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
Continuing Grant
RARE-1: Understanding the Physics of Human Whistling
RARE-1:了解人类口哨的物理原理
- 批准号:
2332390 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
Standard Grant
RII Track-4:NSF:Understanding the Fundamental Physics of Acousto-Magnetic Microswimmers to Realize Precise, Tunable Motion at Microscales
RII Track-4:NSF:了解声磁微型游泳器的基础物理学,以实现微尺度的精确、可调运动
- 批准号:
2229636 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
Standard Grant
Collaborative Research: NSFGEO-NERC: Understanding surface-to-bed meltwater pathways across the Greenland Ice Sheet using machine-learning and physics-based models
合作研究:NSFGEO-NERC:使用机器学习和基于物理的模型了解格陵兰冰盖的地表到床层融水路径
- 批准号:
2344690 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
Standard Grant
Understanding the Critical Role of Seasonality for El Niño/Southern Oscillation (ENSO) Variability, Using Empirical Stochastic-Dynamic Models and Physics-Based Coupled Models
使用经验随机动态模型和基于物理的耦合模型了解季节性对厄尔尼诺/南方涛动 (ENSO) 变异性的关键作用
- 批准号:
2311162 - 财政年份:2023
- 资助金额:
$ 38.62万 - 项目类别:
Standard Grant