Cost Oriented Humanoid Robots

International audience

Visual servoing allows to control the motion of a robot using information from its visual sensors to achieve manipulation tasks. In this work we design and implement a robust visual servoing framework for reaching and grasping behaviours for a humanoid service robot with limited control capabilities. Our approach successfully exploits a 5-degrees of freedom manipulator, overcoming the control limitations of the robot while avoiding singularities and stereo vision techniques. Using a single camera, we combine a marker-less model based tracker for the target object, a pattern tracking for the end-effector to deal with the robot’s inaccurate kinematics, and alternate pose based visual servo technique with eye-in-hand and eye-to-hand configurations to achieve a fully functional grasping system. The overall method shows better results for grasping than conventional motion planing and simple inverse kinematics techniques for this robotic morphology, demonstrating a 48.8% of increment in the grasping success rate.

As Large Language Models (LLMs) become increasingly integrated into everyday life as general-purpose multimodal AI systems, their capabilities to simulate human understanding are under examination. This study investigates LLMs’ ability to interpret linguistic pragmatics, which involves context and implied meanings. Using Grice’s communication principles, we evaluated both LLMs (GPT-2, GPT-3, GPT-3.5, GPT-4, and Bard) and human subjects (N = 147) on dialogue-based tasks. Human participants included 71 primarily Serbian students and 76 native English speakers from the United States. Findings revealed that LLMs, particularly GPT-4, outperformed humans. GPT-4 achieved the highest score of 4.80, surpassing the best human score of 4.55. Other LLMs performed well: GPT-3.5 scored 4.10, Bard 3.75, and GPT-3 3.25; GPT-2 had the lowest score of 1.05. The average LLM score was 3.39, exceeding the human cohorts’ averages of 2.80 (Serbian students) and 2.34 (U.S. participants). In the ranking of all 155 subjects (including LLMs and humans), GPT-4 secured the top position, while the best human ranked second. These results highlight significant progress in LLMs’ ability to simulate understanding of linguistic pragmatics. Future studies should confirm these findings with more dialogue-based tasks and diverse participants. This research has important implications for advancing general-purpose AI models in various communication-centered tasks, including potential application in humanoid robots in the future.

The computation by which our brain elaborates fast responses to emotional expressions is currently an active field of brain studies. Previous studies have focused on stimuli taken from everyday life. Here, we investigated event-related potentials in response to happy vs neutral stimuli of human and non-humanoid robots. At the behavioural level, emotion shortened reaction times similarly for robotic and human stimuli. Early P1 wave was enhanced in response to happy compared to neutral expressions for robotic as well as for human stimuli, suggesting that emotion from robots is encoded as early as human emotion expression. Congruent with their lower faceness properties compared to human stimuli, robots elicited a later and lower N170 component than human stimuli. These findings challenge the claim that robots need to present an anthropomorphic aspect to interact with humans. Taken together, such results suggest that the early brain processing of emotional expressions is not bounded to human-like arrangements embodying emotion.

Recent developments in the areas of mobile and humanoid robots have resulted in a marked increase in the potential applications of robotics. In addition, since it is likely that robots will cohabitate and interact with humans in a variety of situations, they will need an external skin covering capable of distinguishing between various stimuli. This study sought to develop skin sensors for robots that were capable of recognizing four stimuli that are frequently encountered in everyday life. We developed a prototype skin sensor system using two types of shock sensors and pressure-sensitive rubber. The results demonstrated that four stimuli could be distinguished based on differences in pressure and acceleration between each stimulus. We describe the development of the skin sensory system, the mechanism employed to distinguish between the different stimuli, and present the results of experiments conducted using the developed system.

The rapid advancement of robotics has enabled the robots to serve physical help for elderly people in everyday life. This paper investigates what types of robot elderly the most willing to go shopping with. We investigate the effect into following two factors: conversation and robot-type. Conversation means the no purpose talking which human often do. For robot-type, we prepare humanoid and cart robot. To investigate the effect of these factors, a field experiment was conducted in a real supermarket where 24 elderly participants shopped with robots. The experimental results revealed that they prefer a conversational humanoid as a shopping assistant partner.

Our society is on the borderline of information era, experiencing a transition towards a robotic one. Humanoid and android robots are entering with a steady pace into our everyday lives taking up roles related to companionship, partnership, wellness, healthcare, and education among others. The fusion of information technology, ubiquitous computing, robotics, and android science has generated the Geminoid Reality. The Geminoid is a teleoperated, connected to a computer network, android robot that works as a duplicate of an existing person. A motion-capture system tracks facial expressions, and head movements of the operator, and transmits them to the robot, overriding at run-time the preprogrammed configurations of the robots actuators. The Geminoid Reality is combining the Visual Reality (users’ and robot’s point of view) with an Augmented one (operator’s point of view) into a new kind of mixed reality involving physical embodiment, and representation, causing the ownership transfer, and blended presence phenomena.

In recent years, machine learning algorithms have been developing rapidly, becoming increasingly powerful tools in decoding physiological and neural signals. The aim of this dissertation is to develop computational tools, and especially machine learning techniques, to identify the most effective methods for feature extraction and classification of these signals. This is particularly challenging due to the highly non-linear, non-stationery, and artifact- and noise-prone nature of these signals. Among basic human-control tasks, reaching and grasping are ubiquitous in everyday life. I investigated different linear and non-linear dimensionality reduction techniques for feature extraction and classification of electromyography (EMG) during a reach-grasp-lift task. The results highlighted the advantages of completely automated feature-learning by Laplacian Eigenmaps over manual feature engineering, especially when combined with classification, to achieve high accuracy with few training samples. The ability to decode and reduce the complexity of EMG could enable new practical applications for EMG in basic science and in the clinic. It could also help design humanoid and other robots. Beyond EMG, a key objective of my dissertation was to decode brain activity during the decision-making processes that lead to voluntary action. This was based on electroencephalography (EEG) and holds the promise for improved brain-computer interfaces (BCIs), particularly related to motor imagery (MI). We developed an end-to-end convolutional neural network with attentional mechanism together with different data augmentation techniques on two benchmark MI datasets. I also collected a new dataset, recorded using high-density EEG, and containing both MI and motor execution (ME) tasks. This enabled us to directly compare the decodability of MI and ME, investigate optimal channel configurations, and much more. In particular, this facilitates the analysis and decoding of MI on the fly—online and in in real time. Another potential use of EEG is measuring brain activity of people who are floating in body-temperature water in a sensory-deprivation-tank float pod. We compared lying in the float pod versus lying in bed (a control condition). And we found differences between the two, especially in the gamma band. More research is required to understand what these findings mean for levels of stress in the float pod.

Humanoids have recently become a popular research platform in the robotics community. Such robots offer various fields for new applications. However, they have several drawbacks compared to wheeled vehicles such as stability problems, limited payload capabilities, violation of the flat world assumption, and they typically provide only very rough odometry information, if at all. In this paper, we investigate the problem of learning accurate grid maps with humanoid robots. We present techniques to deal with some of the above-mentioned difficulties. We describe how an existing approach to the simultaneous localization and mapping (SLAM) problem can be adapted to robustly learn accurate maps with a humanoid equipped with a laser range finder. We present an experiment in which our mapping system builds a highly accurate map with a size of around 20m by 20m using data acquired with a humanoid in our office environment containing two loops. The resulting maps have a similar accuracy as maps built with a wheeled robot. ©2008 IEEE.

Background: Our team has been fighting nosocomial infections since 1991. During our journey, we often ask why people do not wash their hands! Semmelweiss discovered in the 1840s that handwashing prevented deaths from puerperal sepsis, but we still need to convince healthcare workers about hand hygiene. One answer is that washing hands is an unsophisticated gesture, without any technology, so people just do not do it. How can we improve compliance with hand hygiene? We imagined a robot in our team to remind people to wash their hands. Then, in 2016 we met Meccanoid, a US$200 toy robot: a 4-foot-tall programmable humanoid robot with voice recognition capabilities. We made adaptions in the robot (mini-projector + audio amplifier + alcohol dispenser + spy camera), and we gave him a name (Ozires) and a purpose: He became a professor who teaches healthcare workers how, when, and why wash their hands! Here, we describe the multimodal strategy centered around Ozires. Methods: The multimodal strategy consists of 7 key elements: (1) the robot, accompanied by a infection control practitioner, performs audio and video lectures about hand hygiene techniques, motivational videos, data feedback; (2) the robot’s wood copies with sound alert with motion detector for hand hygiene are spread out in the whole hospital; (3) fridge magnet with robot prints (gifts for patients and healthcare professionals); (4) app for hand hygiene monitoring (Hands Clean); (5) adherence rates by professional category and individual feedback; (6) patient empowerment for hand hygiene; and (7) sound alert for hand hygiene in the patient room’s door. Results: After the insertion of Ozires in 3 ICUs of hospital A (pilot study), the hand hygiene (HH) rate increased from ~36%, between January and July 2016, to ~68% between August 2016 and October 2019. At hospital B, Ozires started his lectures in May 2018, throughout the hospital. Hand hygiene adherence increased from 23% between July and December 2017 to 60% between June 2018 and October 2019. In the 3 months before this multimodal strategy was implemented in hospital C (June–August 2019), and the mean rate of hand hygiene was 65%. With the robot, the hand hygiene rate increased to 94% (September–October 2019). Conclusions: The multimodal strategy centered around the robot Ozires works! Hand hygiene compliance increased significantly after the interventions. People listen the robot much more attentively than to their human colleagues, and healthcare worker behavior changed! We need to go further improve the program, but it is sustainable. Finally, we succeeded in convincing people to improve their hand hygiene practices.Funding: NoneDisclosures: None

Introduction A growing body of research has demonstrated the positive effects of Acceptance and Commitment Therapy (ACT) delivered in the workplace on employees’ well-being and psychological flexibility. Delivering ACT in the workplace aims to enhance employee performance and reduce stress levels by promoting psychological flexibility, a skill to consciously live in the present, observing thoughts and emotions, and engaging in value-driven actions. Objectives The short-term effects of a 5-week ACT-based psychological flexibility training program were tested. Methods Participants were 21 corporate employees (62 % women, M age = 43.9 years, SD = 9.5) recruited from a large company in the financial sector. The six core processes of psychological flexibility were examined before the first training session and after the fifth training session of the 5-week training. Psychological flexibility, value-driven living, thought suppression, five facets of mindfulness and cognitive fusion were assessed. Results Results showed a statistically significant difference in one core process of psychological flexibility, namely Contact with the present moment. Acting with awareness , Describing and Nonjudge showed significant difference before and after the training. Presumably, changes in Cognitive defusion and Non-reactivity are also evident, however results did not show significant difference. Conclusions The present research has provided further empirical evidence for the effectiveness of ACT-based psychological flexibility training with regard to three aspects of Contact with the present moment , which is a core process of psychological flexibility ( Acting with awareness, Describing, and Nonjudge ). The findings provide a basis for future research to investigate longer-term effects, including monitoring how participants in the research program practice the learnt techniques in their everyday lives. Disclosure of Interest Á. Magyary Grant / Research support from: ÁM was supported by the ÚNKP-23-2 New National Excellence Program of the Ministry for Culture and Innovation from the source of the National Research, Development and Innovation Fund., Z. Demetrovics: None Declared, Á. Zsila: None Declared, N. Mészáros: None Declared

The sustainability topic is present in our everyday life. No matter if we are traveling in the public transport, we are having coffee with friends or watching tv we hear something related to green economy, sustainable usage of resources or smart energy saving devices. All ads that we see around us, in one way or another, also are calling to be environmentally friendly and sustainable. On national and international level, there are a lot of regulations and requirements developed and enforced both for the business and for the citizens. In the general case, it is commonly accepted, that the companies are taking actions in the field of sustainable development only because they are forced to do so by legislative requirements and penalties. But is it always the case? The paper discusses the main idea standing behind the sustainable development and presents some of the models that summarizes this idea. Further, by reviewing the practices of three large companies in different economic fields, it discusses their approach and attitude towards the subject.

Research background: Digitalization in the context of the COVID-19 pandemic has accelerated significantly across Europe, including in regions that are at the stage of catch-up development. However, as innovative technologies are intensively integrated into everyday life, the digital gap between the EU countries is increasing. The widening digital gap is becoming a serious threat to achieving the EU's sustainable development goals and building a sustainable European society. Purpose of the article: The purpose of the article was to empirically substantiate the inclusiveness and convergence of the digital strategy in the EU countries. Methods: Using the method of fuzzy sets, the level of asymmetry in access to transactional (e-commerce), information (cloud computing), operational (artificial intelligence) technologies in the EU countries was assessed. The negative impact of the digital gap within the countries (the gap between small, medium, large companies) and the global digital gap (the gap between EU countries) on the competitiveness of countries was established, for which correlation analysis and the Granger causality test were used. Findings & value added: The findings of this study contribute to the literature of digital transformation and digital gap of European countries. The impact of the digital gap in transactional, information, operational technologies on the competitiveness of countries is differentiated depending on the level of digitalization of the country. Cluster groups of countries are determined by the convergence of digitalization and ways to ensure long-term competitiveness. The directions for reducing the digital gap in the EU countries are substantiated by changing the priorities of spending on innovation and increasing productivity, diversifying the digital technologies used. The empirical results obtained can serve as a basis for improving the effectiveness of the digitalization policy in the EU countries in accordance with individual convergence goals. The main added value of the paper is related to the presented research procedure, which can be used in analyses of digital technologies development also for other countries. The results provide valuable insights into evaluating the digital technologies in European countries.

This paper examines the effects on UK audit market concentration and pricing of mergers between the large audit firms and the demise of Andersen. Based on data over the period 1985–2002, it appears that mergers contributed to a rise in concentration ratios to levels that suggest concern about the potential for monopoly pricing. The high concentration ratios have not improved the level of price competition in the UK audit market. Our pooled models suggest that concentration ratios are associated with higher audit fees. The evidence suggests that the effects of mergers between big firms on brand name fee premium and on price competition vary depending on the particular circumstances. The brand name premium is strongest for the largest quartile of companies prior to the mergers. After the Big Six mergers, the premium increases for average‐sized companies but falls for the smallest and largest companies. Following the PricewaterhouseCoopers merger, the premium increases for below median‐sized clients but decreases for above‐median sized clients. For the Deloitte‐Andersen transaction, the premium falls for the smallest and largest companies but increases for those in the second quartile. Our results provide evidence that auditees are likely to pay higher fees if their auditor merges with a larger counterpart. We attribute merger‐related fee hikes to product differentiation, rather than anti‐competitive pricing.

Wood is a building material that is frequently used as the main component in interior design and architecture. But, the use of wood is declining today and it is replaced by bamboo. Now, wood material is difficult to find and this condition causes high price in market. Article presents the availability of bamboo that has been known by our society because of its benefit. They use bamboo to meet their everyday life, such as clothes, food, and house needs. Bamboo is also used for the development of interior design as well as architecture , especially in Indonesia.