In robotics, methods and softwares usually require optimizations of hyperparameters in order to be efficient for specific tasks, for instance industrial bin-picking from homogeneous heaps of different objects. We present a developmental framework based on long-term memory and reasoning modules (Bayesian Optimisation, visual similarity and parameters bounds reduction) allowing a robot to use meta-learning mechanism increasing the efficiency of such continuous and constrained parameters optimizations. The new optimization, viewed as a learning for the robot, can take advantage of past experiences (stored in the episodic and procedural memories) to shrink the search space by using reduced parameters bounds computed from the best optimizations realized by the robot with similar tasks of the new one (e.g. bin-picking from an homogenous heap of a similar object, based on visual similarity of objects stored in the semantic memory). As example, we have confronted the system to the constrained optimizations of 9 continuous hyperparameters for a professional software (Kamido) in industrial robotic arm bin-picking tasks, a step that is needed each time to handle correctly new object. We used a simulator to create bin-picking tasks for 8 different objects (7 in simulation and one with real setup, without and with meta-learning with experiences coming from other similar objects) achieving goods results despite a very small optimization budget, with a better performance reached when meta-learning is used (84.3% vs 78.9% of success overall, with a small budget of 30 iterations for each optimization) for every object tested (p-value=0.036).
Ritsumeikan University（立命馆大学）；Keio University（庆应义塾大学）；The University of Tokyo（东京大学）；Università Milano-Bicocca（米兰比可卡大学）；University of Essex（埃塞克斯大学）；Donders Institute for Brain, Cognition and Behaviour（唐德斯脑、认知与行为研究所）；Cajal International Neuroscience Center, Spanish National Research Council（西班牙国家研究委员会卡哈尔国际神经科学中心）；Bogazici University（博阿齐奇大学）；Queen Mary University of London（伦敦玛丽女王大学）；The University of Electro-Communications（电气通信大学）；National Autonomous University of Mexico（墨西哥国立自治大学）；Universidad Autónoma del Estado de Morelos（莫雷洛斯州自治大学）；Institute of Cognitive Sciences and Technologies, National Research Council of Italy（意大利国家研究委员会认知科学与技术研究所）
Creating autonomous robots that can actively explore the environment, acquire knowledge and learn skills continuously is the ultimate achievement envisioned in cognitive and developmental robotics. Their learning processes should be based on interactions with their physical and social world in the manner of human learning and cognitive development. Based on this context, in this paper, we focus on the two concepts of world models and predictive coding. Recently, world models have attracted renewed attention as a topic of considerable interest in artificial intelligence. Cognitive systems learn…
Even with impressive advances in application-specific models, we still lack knowledge about how to build a model that can learn in a human-like way and do multiple tasks. To learn in a human-like way, we need to provide a diverse experience that is comparable to humans. In this paper, we introduce our ongoing effort to build a simulated environment for developmental robotics (SEDRo). SEDRo provides diverse human experiences ranging from those of a fetus to a 12th-month-old. A series of simulated tests based on developmental psychology will be used to evaluate the progress of a learning model. …
Despite the recent advancement in the social robotic field, important limitations restrain its progress and delay the application of robots in everyday scenarios. In the present paper, we propose to develop computational models inspired by our knowledge of human infants' social adaptive abilities. We believe this may provide solutions at an architectural level to overcome the limits of current systems. Specifically, we present the functional advantages that adaptive Theory of Mind (ToM) systems would support in robotics (i.e., mentalizing for belief understanding, proactivity and preparation, …
A3RNN: Bi-directional Fusion of Bottom-up and Top-down Process for Developmental Visual Attention in Robots
Department of Intermedia Art and Science, Waseda University（媒体艺术与科学系，早稻田大学）；Research Development Group, Hitachi, Ltd.（日立株式会社研发发展组）；National Institute of Advanced Industrial Science and Technology (AIST)（国家先进工业科学与技术研究院）
LIRIS, CNRS UMR 5205（里昂图像与信息系统实验室，法国国家科学研究中心 UMR 5205）；Ecole Centrale de Lyon（里昂中央理工学院）
We present a developmental framework based on a long-term memory and reasoning mechanisms (Vision Similarity and Bayesian Optimisation). This architecture allows a robot to optimize autonomously hyper-parameters that need to be tuned from any action and/or vision module, treated as a black-box. The learning can take advantage of past experiences (stored in the episodic and procedural memories) in order to warm-start the exploration using a set of hyper-parameters previously optimized from objects similar to the new unknown one (stored in a semantic memory). As example, the system has been used…
Task-specific AI agents are showing remarkable performance across different domains. But modeling generalized AI agents like human intelligence will require more than current datasets or only reward-based environments that don't include experiences that an infant gathers throughout its initial stages. In this paper, we present Simulated Environment for Developmental Robotics (SEDRo). It simulates the environments for a baby agent that a human baby experiences throughout the pre-born fetus stage to post-birth 12 months. SEDRo also includes a mother character to provide social interaction with t…
Ritsumeikan University（立命馆大学）；The University of Tokyo（东京大学）；The Whole Brain Architecture Initiative（全脑架构倡议）；Osaka University（大阪大学）；Okinawa Institute of Science and Technology Graduate University（冲绳科学技术大学院大学）；Tamagawa University（玉川大学）；The University of Electro-Communications（电气通信大学）
Building a humanlike integrative artificial cognitive system, that is, an artificial general intelligence (AGI), is the holy grail of the artificial intelligence (AI) field. Furthermore, a computational model that enables an artificial system to achieve cognitive development will be an excellent reference for brain and cognitive science. This paper describes an approach to develop a cognitive architecture by integrating elemental cognitive modules to enable the training of the modules as a whole. This approach is based on two ideas: (1) brain-inspired AI, learning human brain architecture to b…
Ritsumeikan University（立命馆大学）；The University of Electro-Communications（电气通信大学）；Okayama Prefectural University（冈山县立大学）；Waseda University（早稻田大学）；National Institute of Advanced Industrial Science and Technology（产业技术综合研究所）
Humans can learn the use of language through physical interaction with their environment and semiotic communication with other people. It is very important to obtain a computational understanding of how humans can form a symbol system and obtain semiotic skills through their autonomous mental development. Recently, many studies have been conducted on the construction of robotic systems and machine-learning methods that can learn the use of language through embodied multimodal interaction with their environment and other systems. Understanding human social interactions and developing a robot th…
Donders Institute for Brain, Cognition, and Behaviour（唐德斯大脑、认知与行为研究所）；Radboud University（拉德堡德大学）；Joint Research Centre, European Commission（欧盟委员会联合研究中心）；Ghent University（根特大学）；imec（微电子研究中心）；Instituto Superio Técnico（高等技术学院）；University of Lisbon（里斯本大学）
Understanding and modelling children's cognitive processes and their behaviour in the context of their interaction with robots and social artificial intelligence systems is a fundamental prerequisite for meaningful and effective robot interventions. However, children's development involve complex faculties such as exploration, creativity and curiosity which are challenging to model. Also, often children express themselves in a playful way which is different from a typical adult behaviour. Different children also have different needs, and it remains a challenge in the current state of the art t…
Developmental psychology and neuroimaging research identified a close link between numbers and fingers, which can boost the initial number knowledge in children. Recent evidence shows that a simulation of the children's embodied strategies can improve the machine intelligence too. This article explores the application of embodied strategies to convolutional neural network models in the context of developmental neuro-robotics, where the training information is likely to be gradually acquired while operating rather than being abundant and fully available as the classical machine learning scenari…
Tufts University（塔夫茨大学）；Wyss Institute for Biologically Inspired Engineering（怀斯生物启发工程研究所）；Harvard University（哈佛大学）；University of Vermont（佛蒙特大学）；Institute for Computationally Designed Organisms（计算设计生物体研究所）
Advances in science and engineering often reveal the limitations of classical approaches initially used to understand, predict, and control phenomena. With progress, conceptual categories must often be re-evaluated to better track recently discovered invariants across disciplines. It is essential to refine frameworks and resolve conflicting boundaries between disciplines such that they better facilitate, not restrict, experimental approaches and capabilities. In this essay, we discuss issues at the intersection of developmental biology, computer science, and robotics. In the context of biologi…
Advances in Artificial Intelligence and robotics are currently questioning theethical framework of their applications to deal with potential drifts, as well as the way inwhich these algorithms learn because they will have a strong impact on the behavior ofrobots and the type of robots. interactions with people. We would like to highlight someprinciples and ideas from cognitive neuroscience and development sciences based on theimportance of the body for intelligence, contrary to the theory of the all-brain or all-algorithm, to represent the world and interacting with others, and their current a…
A key challenge in Imitation Learning (IL) is that optimal state actions demonstrations are difficult for the teacher to provide. For example in robotics, providing kinesthetic demonstrations on a robotic manipulator requires the teacher to control multiple degrees of freedom at once. The difficulty of requiring optimal state action demonstrations limits the space of problems where the teacher can provide quality feedback. As an alternative to state action demonstrations, the teacher can provide corrective feedback such as their preferences or rewards. Prior work has created algorithms designe…
A large body of compelling evidence has been accumulated demonstrating that embodiment - the agent's physical setup, including its shape, materials, sensors and actuators - is constitutive for any form of cognition and as a consequence, models of cognition need to be embodied. In contrast to methods from empirical sciences to study cognition, robots can be freely manipulated and virtually all key variables of their embodiment and control programs can be systematically varied. As such, they provide an extremely powerful tool of investigation. We present a robotic bottom-up or developmental appr…
Okinawa Institute of Science and Technology（冲绳科学技术大学院大学）
Evolution and development operate at different timescales; generations for the one, a lifetime for the other. These two processes, the basis of much of life on earth, interact in many non-trivial ways, but their temporal hierarchy -- evolution overarching development -- is observed for most multicellular lifeforms. When designing robots however, this tenet lifts: it becomes -- however natural -- a design choice. We propose to inverse this temporal hierarchy and design a developmental process happening at the phylogenetic timescale. Over a classic evolutionary search aimed at finding good gaits…
CNRS（法国国家科学研究中心）；ISIR（智能系统与机器人研究所）；U2IS（U2IS实验室）；INRIA Flowers（法国国家信息与自动化研究所Flowers团队）；ENSTA（法国国立高等先进技术学院）；Institut Polytechnique Paris（巴黎理工学院）；University of Edinburgh（爱丁堡大学）
Robots are still limited to controlled conditions, that the robot designer knows with enough details to endow the robot with the appropriate models or behaviors. Learning algorithms add some flexibility with the ability to discover the appropriate behavior given either some demonstrations or a reward to guide its exploration with a reinforcement learning algorithm. Reinforcement learning algorithms rely on the definition of state and action spaces that define reachable behaviors. Their adaptation capability critically depends on the representations of these spaces: small and discrete spaces re…
Long-term autonomy of robotic systems implicitly requires dependable platforms that are able to naturally handle hardware and software faults, problems in behaviors, or lack of knowledge. Model-based dependable platforms additionally require the application of rigorous methodologies during the system development, including the use of correct-by-construction techniques to implement robot behaviors. As the level of autonomy in robots increases, so do the cost of offering guarantees about the dependability of the system. Certifiable dependability of autonomous robots, we argue, can benefit from f…
Sensorimotor contingency theory offers a promising account of the nature of perception, a topic rarely addressed in the robotics community. We propose a developmental framework to address the problem of the autonomous acquisition of sensorimotor contingencies by a naive robot. While exploring the world, the robot internally encodes contingencies as predictive models that capture the structure they imply in its sensorimotor experience. Three preliminary applications are presented to illustrate our approach to the acquisition of perceptive abilities: discovering the environment, discovering obje…
Toward General-Purpose Robots via Foundation Models: A Survey and Meta-Analysis
CMU（卡内基梅隆大学）；Bosch Center for AI（博世人工智能中心）；MIT（麻省理工学院）；Field AI（菲尔德人工智能公司）；Georgia Tech（佐治亚理工学院）；FAIR at Meta（元公司人工智能研究院）；UC San Diego（加利福尼亚大学圣迭戈分校）；SAIR Lab（空间人工智能与机器人实验室）；Google DeepMind（谷歌DeepMind）
Orthogonal Research and Education Laboratory（正交研究与教育实验室）；OpenWorm Foundation（开放蠕虫基金会）
What role does phenotypic complexity play in the systems-level function of an embodied agent? The organismal phenotype is a topologically complex structure that interacts with a genotype, developmental physics, and an informational environment. Using this observation as inspiration, we utilize a type of embodied agent that exhibits layered representational capacity: meta-brain models. Meta-brains are used to demonstrate how phenotypes process information and exhibit self-regulation from development to maturity. We focus on two candidate structures that potentially explain this capacity: foldin…
Understanding infant development is one of the greatest scientific challenges of contemporary science. A large source of difficulty comes from the fact that the development of skills in infants results from the interactions of multiple mechanisms at multiple spatio-temporal scales. The concepts of "innate" or "acquired" are not any more adequate tools for explanations, which call for a shift from reductionist to systemic accounts. To address this challenge, building and experimenting with robots modeling the growing infant brain and body is crucial. Systemic explanations of pattern formation i…
University of Plymouth（普利茅斯大学）；University of Manchester（曼彻斯特大学）；AIST-AIRC（日本产业技术综合研究所人工智能研究中心）；Sheffield Hallam University（谢菲尔德哈勒姆大学）
In this paper, a novel neuro-robotics model capable of counting real items is introduced. The model allows us to investigate the interaction between embodiment and numerical cognition. This is composed of a deep neural network capable of image processing and sequential tasks performance, and a robotic platform providing the embodiment - the iCub humanoid robot. The network is trained using images from the robot's cameras and proprioceptive signals from its joints. The trained model is able to count a set of items and at the same time points to them. We investigate the influence of pointing on …
University of California, San Diego（加利福尼亚大学圣地亚哥分校）
This paper presents an approach to exploring a multi-objective reinforcement learning problem with Model-Agnostic Meta-Learning. The environment we used consists of a 2D vehicle equipped with a LIDAR sensor. The goal of the environment is to reach some pre-determined target location but also effectively avoid any obstacles it may find along its path. We also compare this approach against a baseline TD3 solution that attempts to solve the same problem.
Inria（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；Université de Lorraine（洛林大学）
Meta-learning algorithms can accelerate the model-based reinforcement learning (MBRL) algorithms by finding an initial set of parameters for the dynamical model such that the model can be trained to match the actual dynamics of the system with only a few data-points. However, in the real world, a robot might encounter any situation starting from motor failures to finding itself in a rocky terrain where the dynamics of the robot can be significantly different from one another. In this paper, first, we show that when meta-training situations (the prior situations) have such diverse dynamics, usi…
Stanford University（斯坦福大学）；KTH Royal Institute of Technology（瑞典皇家理工学院）
Reinforcement learning methods can achieve significant performance but require a large amount of training data collected on the same robotic platform. A policy trained with expensive data is rendered useless after making even a minor change to the robot hardware. In this paper, we address the challenging problem of adapting a policy, trained to perform a task, to a novel robotic hardware platform given only few demonstrations of robot motion trajectories on the target robot. We formulate it as a few-shot meta-learning problem where the goal is to find a meta-model that captures the common stru…
Robotic insertion tasks are characterized by contact and friction mechanics, making them challenging for conventional feedback control methods due to unmodeled physical effects. Reinforcement learning (RL) is a promising approach for learning control policies in such settings. However, RL can be unsafe during exploration and might require a large amount of real-world training data, which is expensive to collect. In this paper, we study how to use meta-reinforcement learning to solve the bulk of the problem in simulation by solving a family of simulated industrial insertion tasks and then adapt…
University of California Berkeley（加州大学伯克利分校）；Stanford University（斯坦福大学）
Meta-reinforcement learning algorithms can enable autonomous agents, such as robots, to quickly acquire new behaviors by leveraging prior experience in a set of related training tasks. However, the onerous data requirements of meta-training compounded with the challenge of learning from sensory inputs such as images have made meta-RL challenging to apply to real robotic systems. Latent state models, which learn compact state representations from a sequence of observations, can accelerate representation learning from visual inputs. In this paper, we leverage the perspective of meta-learning as …
Okinawa Institute of Science and Technology Graduate University（冲绳科学技术大学院大学）
Human-robot interaction is becoming an interesting area of research in cognitive science, notably, for the study of social cognition. Interaction theorists consider primary intersubjectivity a non-mentalist, pre-theoretical, non-conceptual sort of processes that ground a certain level of communication and understanding, and provide support to higher-level cognitive skills. We argue this sort of low level cognitive interaction, where control is shared in dyadic encounters, is susceptible of study with neural robots. Hence, in this work we pursue three main objectives. Firstly, from the concept …
University of Hamburg（汉堡大学）；Humboldt-Universität zu Berlin（柏林洪堡大学）；University of Potsdam（波茨坦大学）；Max Planck Institute for Human Cognitive and Brain Sciences（马克斯·普朗克人类认知与脑科学研究所）
Safe human-robot interactions require robots to be able to learn how to behave appropriately in \sout{humans' world} \rev{spaces populated by people} and thus to cope with the challenges posed by our dynamic and unstructured environment, rather than being provided a rigid set of rules for operations. In humans, these capabilities are thought to be related to our ability to perceive our body in space, sensing the location of our limbs during movement, being aware of other objects and agents, and controlling our body parts to interact with them intentionally. Toward the next generation of robots…
University of Waterloo（滑铁卢大学）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）
There has been increasing interest in smart factories powered by robotics systems to tackle repetitive, laborious tasks. One impactful yet challenging task in robotics-powered smart factory applications is robotic grasping: using robotic arms to grasp objects autonomously in different settings. Robotic grasping requires a variety of computer vision tasks such as object detection, segmentation, grasp prediction, pick planning, etc. While significant progress has been made in leveraging of machine learning for robotic grasping, particularly with deep learning, a big challenge remains in the need…
University of Science and Technology of China（中国科学技术大学）；Cyberspace Institute of Advanced Technology, Guangzhou University（广州大学网络空间先进技术研究院）
Traditionally, reinforcement learning methods predict the next action based on the current state. However, in many situations, directly applying actions to control systems or robots is dangerous and may lead to unexpected behaviors because action is rather low-level. In this paper, we propose a novel hierarchical reinforcement learning framework without explicit action. Our meta policy tries to manipulate the next optimal state and actual action is produced by the inverse dynamics model. To stabilize the training process, we integrate adversarial learning and information bottleneck into our fr…
Time to Play: Simulating Early-Life Animal Dynamics Enhances Robotics Locomotion Discovery
Adaptive & Intelligent Robotics Lab, Imperial College London, United Kingdom（智能适应机器人实验室，帝国理工学院伦敦分校）；Evolutionary Biomechanics Laboratory, Imperial College London, United Kingdom（进化生物力学实验室，帝国理工学院伦敦分校）
Training a reinforcement learning (RL) agent on a real-world robotics task remains generally impractical due to sample inefficiency. Multi-task RL and meta-RL aim to improve sample efficiency by generalizing over a distribution of related tasks. However, doing so is difficult in practice: In multi-task RL, state of the art methods often fail to outperform a degenerate solution that simply learns each task separately. Hypernetworks are a promising path forward since they replicate the separate policies of the degenerate solution while also allowing for generalization across tasks, and are appli…
The act of reaching for an object is a fundamental yet complex skill for a robotic agent, requiring a high degree of visuomotor control and coordination. In consideration of dynamic environments, a robot capable of autonomously adapting to novel situations is desired. In this paper, a developmental robotics approach is used to autonomously learn visuomotor coordination on the NICO (Neuro-Inspired COmpanion) platform, for the task of object reaching. The robot interacts with its environment and learns associations between motor commands and temporally correlated sensory perceptions based on Heb…
Meta-Learning for Fast Adaptation in Intent Inferral on a Robotic Hand Orthosis for Stroke
Department of Mechanical Engineering, Columbia University in the City of New York（机械工程系，哥伦比亚大学（纽约市））；Department of Computer Science, Columbia University in the City of New York（计算机科学系，哥伦比亚大学（纽约市））；Department of Rehabilitation and Regenerative Medicine, Columbia University Irving Medical Center（康复与再生医学系，哥伦比亚大学伊万斯医学中心）
Soft robots are typically approximated as low-dimensional systems, especially when learning-based methods are used. This leads to models that are limited in their capability to predict the large number of deformation modes and interactions that a soft robot can have. In this work, we present a deep-learning methodology to learn high-dimensional visual models of a soft robot combining multimodal sensorimotor information. The models are learned in an end-to-end fashion, thereby requiring no intermediate sensor processing or grounding of data. The capabilities and advantages of such a modelling a…
MetaTune: Adjoint-based Meta-tuning via Robotic Differentiable Dynamics
School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（华中科技大学人工智能与自动化学院）；School of Electrical and Computer Engineering, National University of Singapore（新加坡国立大学电气与计算机工程系）；The Shenzhen Huazhong University of Science and Technology Research Institute（华中科技大学深圳研究院）
In machine learning, meta-learning methods aim for fast adaptability to unknown tasks using prior knowledge. Model-based meta-reinforcement learning combines reinforcement learning via world models with Meta Reinforcement Learning (MRL) for increased sample efficiency. However, adaption to unknown tasks does not always result in preferable agent behavior. This paper introduces a new Meta Adaptation Controller (MAC) that employs MRL to apply a preferred robot behavior from one task to many similar tasks. To do this, MAC aims to find actions an agent has to take in a new task to reach a similar …
Dynamics-Aware Meta-Imitation for Generalization to Unseen Robotic Manipulation
State Key Laboratory of Robotics and Intelligent Systems, Shenyang Institute of Automation, Chinese Academy of Sciences（中国科学院沈阳自动化研究所机器人学国家重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）；Shenyang University of Technology（沈阳工业大学）
Developmental Robotics offers a new approach to numerous AI features that are often taken as granted. Traditionally, perception is supposed to be an inherent capacity of the agent. Moreover, it largely relies on models built by the system's designer. A new approach is to consider perception as an experimentally acquired ability that is learned exclusively through the analysis of the agent's sensorimotor flow. Previous works, based on H.Poincaré's intuitions and the sensorimotor contingencies theory, allow a simulated agent to extract the dimension of geometrical space in which it is immersed w…
This paper presents a combined task and motion planner for a robot arm to carry out 3D metal wire curving tasks by collaborating with a bending machine. We assume a collaborative robot that is safe to work in a human environment but has a weak payload to bend objects with large stiffness, and developed a combined planner for the robot to use a bending machine. Our method converts a 3D curve to a bending set and generates the feasible bending sequence, machine usage, robotic grasp poses, and pick-and-place arm motion considering the combined task and motion level constraints. Compared with prev…
MIT Computer Science and Artificial Intelligence Laboratory（麻省理工学院计算机科学与人工智能实验室）
Many prediction problems, such as those that arise in the context of robotics, have a simplifying underlying structure that, if known, could accelerate learning. In this paper, we present a strategy for learning a set of neural network modules that can be combined in different ways. We train different modular structures on a set of related tasks and generalize to new tasks by composing the learned modules in new ways. By reusing modules to generalize we achieve combinatorial generalization, akin to the "infinite use of finite means" displayed in language. Finally, we show this improves perform…
