Manipulation Actions Research Articles

Attentional coordination in demonstrator-observer dyads facilitates learning and predicts performance in a novel manual task

Observational learning is a form of social learning in which a demonstrator performs a target task in the company of an observer, who may as a consequence learn something about it. In this study, we approach social learning in terms of the dynamics of coordination rather than the more common perspective of transmission of information. We hypothesised that observers must continuously adjust their visual attention relative to the demonstrator's time-evolving behaviour to benefit from it. We eye-tracked observers repeatedly watching videos showing a demonstrator solving one of three manipulative puzzles before attempting at the task. The presence of the demonstrator's face and the availability of his verbal instruction in the videos were manipulated. We then used recurrence quantification analysis to measure the dynamics of coordination between the overt attention of the observers and the demonstrator's manipulative actions. Bayesian hierarchical logistic regression was applied to examine (1) whether the observers' performance was predicted by such indexes of coordination, (2) how performance changed as they accumulated experience, and (3) if the availability of speech and intentional gaze of the demonstrator mediated it. Results showed that learners better able to coordinate their eye movements with the manipulative actions of the demonstrator had an increasingly higher probability of success in solving the task. The availability of speech was beneficial to learning, whereas the presence of the demonstrator's face was not. We argue that focusing on the dynamics of coordination between individuals may greatly improve understanding of the cognitive processes underlying social learning.

Endocannabinoid System Components as Potential Biomarkers in Psychiatry

The high heterogeneity of psychiatric disorders leads to a lack of diagnostic precision. Therefore, the search of biomarkers is a fundamental aspect in psychiatry to reach a more personalized medicine. The endocannabinoid system (ECS) has gained increasing interest due to its involvement in many different functional processes in the brain, including the regulation of emotions, motivation, and cognition. This article reviews the role of the main components of the ECS as biomarkers in certain psychiatric disorders. Studies carried out in rodents evaluating the effects of pharmacological and genetic manipulation of cannabinoid receptors or endocannabinoids (eCBs) degrading enzymes were included. Likewise, the ECS-related alterations occurring at the molecular level in animal models reproducing some behavioral and/or neuropathological aspects of psychiatric disorders were reviewed. Furthermore, clinical studies evaluating gene or protein alterations in post-mortem brain tissue or in vivo blood, plasma, and cerebrospinal fluid (CSF) samples were analyzed. Also, the results from neuroimaging studies using positron emission tomography (PET) or functional magnetic resonance (fMRI) were included. This review shows the close involvement of cannabinoid receptor 1 (CB1r) in stress regulation and the development of mood disorders [anxiety, depression, bipolar disorder (BD)], in post-traumatic stress disorder (PTSD), as well as in the etiopathogenesis of schizophrenia, attention deficit hyperactivity disorder (ADHD), or eating disorders (i.e. anorexia and bulimia nervosa). On the other hand, recent results reveal the potential therapeutic action of the endocannabinoid tone manipulation by inhibition of eCBs degrading enzymes, as well as by the modulation of cannabinoid receptor 2 (CB2r) activity on anxiolytic, antidepressive, or antipsychotic associated effects. Further clinical research studies are needed; however, current evidence suggests that the components of the ECS may become promising biomarkers in psychiatry to improve, at least in part, the diagnosis and pharmacological treatment of psychiatric disorders.

Helping a robot to understand human actions and objects: a grammatical view

Humans are able to perform a wide variety of complex actions manipulating a very large number of objects. We can make predictions on the outcome of our actions and on how to use different objects. Hence, we have excellent action and object understanding. Artificial agents, on the other hand, still miserably fail in this respect. It is particularly puzzling how inexperienced, young humans can acquire such knowledge; bootstrapped by exploration and extended by supervision. In this study we have, therefore, addressed the question how to structure the realm of actions and objects into dynamic representations, which allow for the easy learning of different action and object concepts. Performing different manipulation actions on a table top (e.g. the actions of “making a breakfast”), we show with our robots that this will indeed lead to some kind of implicit (un-reflected) understanding of action and object concepts allowing the agent to generalize actions and redefine object uses according to need.

Manipulation and Learning by Peers in Online Marketplaces

In the classic beauty contest story of Morris and Shin (2002), the coordination game is set exogenously in the payoff function of agents. Our paper studies the existence of endogenous coordination in a context with one seller and many buyers. Due to imperfect information, the seller has an incentive to take a manipulative action to bias the product's quality. We show that a direct relationship between the seller and buyers does not induce them to coordinate to understand the product's value better. Coordination may occur if and only if a platform's intermediation is in place, as its recommendation to buyers unravels the beneficial effects of the others' documentation efforts. The platform's service works as a coordination device among peers. We finally model the combination of private and public signals with a platform's recommendation. The platform can have a preference for a peer-review system or an individual learning strategy, depending on the seller's ability to manipulate.

Humans Predict Action using Grammar-like Structures

Efficient action prediction is of central importance for the fluent workflow between humans and equally so for human-robot interaction. To achieve prediction, actions can be algorithmically encoded by a series of events, where every event corresponds to a change in a (static or dynamic) relation between some of the objects in the scene. These structures are similar to a context-free grammar and, importantly, within this framework the actual objects are irrelevant for prediction, only their relational changes matter. Manipulation actions and others can be uniquely encoded this way. Using a virtual reality setup and testing several different manipulation actions, here we show that humans predict actions in an event-based manner following the sequence of relational changes. Testing this with chained actions, we measure the percentage predictive temporal gain for humans and compare it to action-chains performed by robots showing that the gain is approximately equal. Event-based and, thus, object independent action recognition and prediction may be important for cognitively deducing properties of unknown objects seen in action, helping to address bootstrapping of object knowledge especially in infants.

Search versus Decision for Election Manipulation Problems

Most theoretical definitions about the complexity of manipulating elections focus on the decision problem of recognizing which instances can be successfully manipulated rather than the search problem of finding the successful manipulative actions. Since the latter is a far more natural goal for manipulators, that definitional focus may be misguided if these two complexities can differ. Our main result is that they probably do differ: If P ≠ NP ∩ coNP (which itself is well known to hold if integer factoring is hard), then for election manipulation, election bribery, and some types of election control, there are election systems for which the problem of recognizing which instances can be successfully manipulated is polynomial-time solvable, yet the task of producing the successful manipulations cannot be done in polynomial time.

Self-Supervised Learning of State Estimation for Manipulating Deformable Linear Objects

We demonstrate model-based, visual robot manipulation of linear deformable objects. Our approach is based on a state-space representation of the physical system that the robot aims to control. This choice has multiple advantages, including the ease of incorporating physics priors in the dynamics model and perception model, and the ease of planning manipulation actions. In addition, physical states can naturally represent object instances of different appearances. Therefore, dynamics in the state space can be learned in one setting and directly used in other visually different settings. This is in contrast to dynamics learned in pixel space or latent space, where generalization to visual differences are not guaranteed. Challenges in taking the state-space approach are the estimation of the high-dimensional state of a deformable object from raw images, where annotations are very expensive on real data, and finding a dynamics model that is both accurate, generalizable, and efficient to compute. We are the first to demonstrate self-supervised training of rope state estimation on real images, without requiring expensive annotations. This is achieved by our novel self-supervising learning objective, which is generalizable across a wide range of visual appearances. With estimated rope states, we train a fast and differentiable neural network dynamics model that encodes the physics of mass-spring systems. Our method has a higher accuracy in predicting future states compared to models that do not involve explicit state estimation and do not use any physics prior, while only using 3\% of training data. We also show that our approach achieves more efficient manipulation, both in simulation and on a real robot, when used within a model predictive controller.

Learning Object-Action Relations from Bimanual Human Demonstration Using Graph Networks

Recognizing human actions is a vital task for a humanoid robot, especially in domains like programming by demonstration. Previous approaches on action recognition primarily focused on the overall prevalent action being executed, but we argue that bimanual human motion cannot always be described sufficiently with a single action label. We present a system for framewise action classification and segmentation in bimanual human demonstrations. The system extracts symbolic spatial object relations from raw RGB-D video data captured from the robot's point of view in order to build graph-based scene representations. To learn object-action relations, a graph network classifier is trained using these representations together with ground truth action labels to predict the action executed by each hand. We evaluated the proposed classifier on a new RGB-D video dataset showing daily action sequences focusing on bimanual manipulation actions. It consists of 6 subjects performing 9 tasks with 10 repetitions each, which leads to 540 video recordings with 2 hours and 18 minutes total playtime and per-hand ground truth action labels for each frame. We show that the classifier is able to reliably identify (action classification macro F1-score of 0.86) the true executed action of each hand within its top 3 predictions on a frame-by-frame basis without prior temporal action segmentation.

Інформаційні виклики для України в умовах російсько-української гібридної війни

Акцентовано на необхідності комплексного аналізу інформаційно-комунікаційної складової частини російсько-українського конфлікту. Досліджено питання інформаційних викликів для України як результату деструктивних пропагандистськоманіпулятивних актів російських владних структур та засобів масової інформації, поширення інформаційної експансії й агресії Російської Федерації. Визначено, що інформаційні виклики мають як короткостроковий, так і довгостроковий виміри. Інформаційний вплив із боку РФ зумовлюється історичною традицією, спільністю розвитку, соціокультурною спадщиною. Окрім того, в основу інформаційної агресії щодо України покладено концепцію «руського миру», що активно популяризується російськими медіа, науковцями та політиками, і становить потенційну загрозу суверенітету нашої держави, збереженню цілісності її території.У процесі формування інформаційних викликів для України визначено три етапи: 1991– 1999 рр. – початковий етап; кінець 2013 – початок 2014 р. – розгортання активних пропагандистсько-маніпулятивних акцій із боку російських спецслужб, політикуму та ЗМІ з метою зміни поглядів, свідомості й вчинків представників українського суспільства, а також створення негативного іміджу України у світі. Третій етап, починаючи з 2014 і до сьогодні, яскраво демонструє масштаби вторгнення РФ в інформаційно-комунікаційний простір України та свідомість українців. Визначено, що збереження інформаційного суверенітету, формування ефективної системи безпеки в інформаційній сфері є актуальною проблемою для України, яка сьогодні є об’єктом зовнішньої інформаційної експансії, маніпулятивних технологій та руйнівного інформаційного вторгнення.

Two-Fingered In-Hand Object Handling Based on Force/Tactile Feedback

This article describes a set of control algorithms for in-hand object handling using a parallel jaw gripper equipped with force/tactile sensors. The control strategy is model based and relies upon the limit surface concept. The LuGre friction model is combined with the limit surface method to set up a dynamic model of soft contact. The model is exploited to estimate the relative velocity of the object with respect to the fingers, so as to control the grip force to counteract possible slipping events due to external disturbances. Force/tactile feedback, the only perception source used by the algorithms, is suitably exploited not only for safe grasping of a variety of objects with uncertain weight and inertial properties, but also for in-hand manipulation actions, like object pivoting or gripper pivoting. Since the algorithm is based on the control of the object velocity, accuracy of the desired object positioning depends on the initial grasp configuration, as well as on the accuracy of the friction model parameters. Such manipulation skills are evaluated in the execution of various pick and place tasks typical of an in-store logistic scenario.

Organization of Frontostriate Interactions with the Involvement of the Brain Error Detector in Supporting Deceptive and Honest Manipulative Actions

We report here studies on the cerebral mechanisms of conscious deceptive and honest manipulative actions. Analysis of cause-effect relationships showed that the execution of both deceptive and honest actions with the aim of manipulating the opinion of an opponent was characterized by a bottom-up interaction between components of the frontostriate system: the caudate nuclei modulated the activity of the prefrontal cortex. This result confirms the hypothesis that the cerebral error detection (ED) mechanism is involved in selection of the appropriate version of an action during execution of manipulative actions.

Cut & recombine: reuse of robot action components based on simple language instructions

Human beings can generalize from one action to similar ones. Robots cannot do this and progress concerning information transfer between robotic actions is slow. We have designed a system that performs action generalization for manipulation actions in different scenarios. It relies on an action representation for which we perform code-snippet replacement, combining information from different actions to form new ones. The system interprets human instructions via a parser using simplified language. It uses action and object names to index action data tables (ADTs), where execution-relevant information is stored. We have created an ADT database from three different sources (KUKA LWR, UR5, and simulation) and show how a new ADT is generated by cutting and recombining data from existing ADTs. To achieve this, a small set of action templates is used. After parsing a new instruction, index-based searching finds similar ADTs in the database. Then the action template of the new action is matched against the information in the similar ADTs. Code snippets are extracted and ranked according to matching quality. The new ADT is created by concatenating code snippets from best matches. For execution, only coordinate transforms are needed to account for the poses of the objects in the new scene. The system was evaluated, without additional error correction, using 45 unknown objects in 81 new action executions, with 80% success. We then extended the method including more detailed shape information, which further reduced errors. This demonstrates that cut & recombine is a viable approach for action generalization in service robotic applications.

From Verb to Action: A Mechanics-based Approach for Trajectory-Pattern Interpretation of Manipulation Verbs

Interpreting manipulation verbs in natural-language instructions into manipulation actions is an indispensable ability for future service robots. As a typical example of manipulation verbs, cutting and breaking (C&B) verbs has gained increasing attention, and various interpretation methods have been proposed in both fields of robotics and natural language processing. However, the existing methods are either relying on the manual meaning matching with tools, objects, and demonstration videos, or based on qualitative and empirical knowledge. Consequently, it is unable to quantitatively interpret the C&B verbs into the robots’ trajectory patterns, which are the necessary information for performing manipulation actions. To address this problem, in this paper, we propose a mechanics-based approach, using the contact, twist, and wrench formulations, for interpreting the C&B verbs into quantitative, computable, and verifiable forms. The proposed method leads to the quantitative constraints and unified validation criteria for robot trajectory generation of manipulation tasks, as well as the semantic analysis in natural language processing.

Mirroring without Overimitation: Learning Functionally Equivalent Manipulation Actions

This paper presents a mirroring approach, inspired by the neuroscience discovery of the mirror neurons, to transfer demonstrated manipulation actions to robots. Designed to address the different embodiments between a human (demonstrator) and a robot, this approach extends the classic robot Learning from Demonstration (LfD) in the following aspects:i) It incorporates fine-grained hand forces collected by a tactile glove in demonstration to learn robot’s fine manipulative actions; ii) Through model-free reinforcement learning and grammar induction, the demonstration is represented by a goal-oriented grammar consisting of goal states and the corresponding forces to reach the states, independent of robot embodiments; iii) A physics-based simulation engine is applied to emulate various robot actions and mirrors the actions that are functionally equivalent to the human’s in the sense of causing the same state changes by exerting similar forces. Through this approach, a robot reasons about which forces to exert and what goals to achieve to generate actions (i.e., mirroring), rather than strictly mimicking demonstration (i.e., overimitation). Thus the embodiment difference between a human and a robot is naturally overcome. In the experiment, we demonstrate the proposed approach by teaching a real Baxter robot with a complex manipulation task involving haptic feedback—opening medicine bottles.

Deep effect trajectory prediction in robot manipulation

Imagining the consequences of one’s own actions, before and during their execution, allows the agents to choose actions based on their simulated performance, and to monitor the progress by comparing observed to simulated behavior. In this study, we propose a deep model that enables a robot to learn to predict the consequences of its manipulation actions from its own interaction experience on objects of various shapes. Given the top-down image of the object, the robot learns to predict the movement trajectory of the object during execution of a lever-up action performed with a screwdriver in a physics-based simulator. The prediction is realized in two stages; the system first computes a number of features from the object and then generates the complete motion trajectory of the center of mass of the object using Long Short Term Memory (LSTM) models. In the first step, we investigated use of various feature descriptors such as shape context that encodes a distributed representation of positions of the object boundary points, unsupervised features that are extracted from autoencoders, Convolutional Neural Network (CNN) based features that are conjointly trained with the LSTMs, and finally task-specific supervised features that are engineered to well-encode the underlying dynamics of the lever-up action. The models are trained in simulation with objects of varying edge numbers and tested in the simulated and the real world. Our deep and generic CNN-based LSTM model outperformed the predictors that use unsupervised representations such as shape descriptors or autoencoder features in the simulated test set. Additionally, it was shown to generalize well to novel object shapes that were not experienced during model training. Finally, our model was shown to perform well in predicting the consequences of lever-up actions generated by a screwdriver that was attached to the gripper of the real UR10 robot. We further showed that our system can predict qualitatively different trajectories of objects that roll off the table or tumble over as the result of lever-up action.

Socio-psychological factors of emotional burnout in personnel of international missions in Ukraine

The syndrome of emotional (professional) burnout is the state of emotional, mental exhaustion, physical exhaustion, which arises as a result of chronic stress at work. It is a result of the interaction of objective stressful environmental impacts and subjective sociopsychological characteristics of an employee. The structure, stages and indicators of burnout syndrome are theoretically analysed. The study of factors contributing to professional burnout was carried out on the staff of the European Union Advisory Mission (EUAM) in Ukraine. It was found that about a quarter of the sample has a high integral index of burnout, which is more often manifested through depersonalization and reduction of professional and personal achievements, but emotional exhaustion is less common in our subjects. The analysis of the relationship between the indicators of emotional professional burnout and some socio-psychological indicators of the subjects in our sample showed that the most important for preventing of burnout occurrence are impulsiveness, practicality and internal intrinsic motivation. Positive correlations with burnout rates (can contribute to its emergence) are the duration of service in missions and such coping strategies as search for social support, cautious, manipulative, antisocial and aggressive actions.

End-state comfort effect in manipulative motor actions of typical and atypical children: a systematic review

ABSTRACT This study aimed to make a systematic review of scientific articles on the planning of manual motor actions of typical and atypical children. To do so, a search of articles published between 1996 and 2017 was done on PubMed, LILACS, Science Direct, and SciElo databases. Original articles in English and Portuguese evaluating the planning of motor actions in typical and atypical children performing manual tasks were selected. For the analysis, the population age, number of children, type of task, main results, site of study and impact of the journal were considered. From the eighteen articles found, twelve were about typical children (from nine months old to twenty years old) and six about atypical children (from three to fourteen years old) diagnosed with autism, hemiplegic cerebral palsy, and developmental coordination disorders. In nine- to ten-year-old typical children, the planning ability of manual motor actions develops over the time and is similar to that of an adult. The atypical children showed motor planning ability lower than that of the typical children, and determining the age when this planning ability is acquired was not possible.

Fazer o fogo fazer:

O artigo consiste em um exercício de tecnologia comparada sobre seis tipos de manipulações do fogo no contexto conservacionista. Parto de pesquisa etnográfica junto a brigadistas e gestores ambientais da Estação Ecológica Serra Geral do Tocantins, uma unidade de conservação localizada na região do Jalapão (Tocantins, Brasil). Baseio-me no tratamento tecnosemiótico das ações manipulatórias, recorrendo à antropologia da ação de André-Georges Haudricourt e dialogando com a filosofia biológica das técnicas de matriz francesa. O objetivo é inserir o tema da tecnicidade do fogo numa abordagem genética, mais interessada nos agenciamentos técnicos que na mera classificação funcional. O texto finaliza qualificando e expandindo os modos de existência do fogo sob manejo para além da predicação ferramental em seu senso utilitário, mas também se esquivando de exaltá-lo como sendo uma ruptura ou inovação tecnocientífica desatrelada de toda uma linhagem técnica da qual ele é parte.

MANIPULATION OF CONSCIOUSNESS AS AN OBJECT OF SOCIAL-PEDAGOGICAL RESEARCH

Abstract. Deploying negative trends in the development of modern society, including informatization, makes the problem of manipulating consciousness at different levels of human communication one of the most relevant. This, in turn, requires the scientific community to create an effective system of socio-psychological protection against manipulative actions in the modern information society. In this article, we studied the meaning of manipulation as a form of covert psychological influence. We defined the criteria on the basis of which one can ground the expediency of distinguishing the manipulative influence over other types of psychological impact. The most considerable characteristic feature of manipulation is the thing that the object of manipulation is not aware of the influence exerted upon him that is determined by veiling of the real aims of the subject of manipulation, who influences not only the sphere of the conscious (mind), but also the sphere of the unconscious (instincts, emotions, and needs). The typical features of manipulation are described: the attitude to an object of manipulation as to a means of achieving one’s own goals; the desire to get a one-sided advantage; the hidden nature of influence; the use of psychological vulnerability; the desire to bring an interaction partner into psychological dependence; the creation of a manipulation recipient’s illusion of independent decision-makings and actions. Manipulative influence differs from its other types by the parameters of morality, awareness, criticism and social autonomy. Manipulation becomes dangerous not only because of the hidden nature of influence, but also, depending on a person’s identity resorting to it, on his/her purpose, aims and beliefs, values, personal properties, social status, etc. Thus, the study of psychosocial factors of persons, inclined to manipulations in interpersonal interaction, is the prospect of our future studies.

Library of actions: Implementing a generic robot execution framework by using manipulation action semantics

When a robot has to imitate an observed action sequence, it must first understand the inherent characteristic features of the individual actions. Such features need to reflect the semantics of the action with a high degree of invariance between different demonstrations of the same action. At the same time the machine needs to be able to execute the action sequence in any appropriate situation. In this study, we introduce a new library of actions, which is a generic framework for executing manipulation actions on robotic systems by combining features that capture action semantics with a framework for execution. We focus on manipulation actions and first create a generic representation consisting of symbolic and sub-symbolic components. To link these two domains we introduce a finite state machine allowing for sequential execution with error handling. The framework is developed from observing humans which provides us with a high degree of grounding. To quantitatively evaluate the scalability of the proposed approach, we conducted a large set of experiments involving different actions performed either individually or sequentially with various types of objects in different scene contexts.