Frontiers in Robotics and AI

RSS Feed for Frontiers in Robotics and AI | New and Recent Articles
Subscribe to Frontiers in Robotics and AI feed

Technologies change rapidly our perception of reality, moving from augmented to virtual to magical. While e-textiles are a key component in exergame or space suits, the transformative potential of the internal side of garments to create embodied experiences still remains largely unexplored. This paper is the result from an art-science collaborative project that combines recent neuroscience findings, body-centered design principles and 2D vibrotactile array-based fabrics to alter one's body perception. We describe an iterative design process intertwined with two user studies on the effects on body-perceptions and emotional responses of various vibration patterns within textile that were designed as spatial haptic metaphors. Our results show potential in considering materials (e.g., rocks) as sensations to design for body perceptions (e.g., being heavy, strong) and emotional responses. We discuss these results in terms of sensory effects on body perception and synergetic impact to research on embodiment in virtual environments, human-computer interaction, and e-textile design. The work brings a new perspective to the sensorial design of embodied experiences which is based on “material perception” and haptic metaphors, and highlights potential opportunities opened by haptic clothing to change body-perception.

The emergence and development of cognitive strategies for the transition from exploratory actions towards intentional problem-solving in children is a key question for the understanding of the development of human cognition. Researchers in developmental psychology have studied cognitive strategies and have highlighted the catalytic role of the social environment. However, it is not yet adequately understood how this capacity emerges and develops in biological systems when they perform a problem-solving task in collaboration with a robotic social agent. This paper presents an empirical study in a human-robot interaction (HRI) setting which investigates children's problem-solving from a developmental perspective. In order to theoretically conceptualize children's developmental process of problem-solving in HRI context, we use principles based on the intuitive theory and we take into consideration existing research on executive functions with a focus on inhibitory control. We considered the paradigm of the Tower of Hanoi and we conducted an HRI behavioral experiment to evaluate task performance. We designed two types of robot interventions, “voluntary” and “turn-taking”—manipulating exclusively the timing of the intervention. Our results indicate that the children who participated in the voluntary interaction setting showed a better performance in the problem solving activity during the evaluation session despite their large variability in the frequency of self-initiated interactions with the robot. Additionally, we present a detailed description of the problem-solving trajectory for a representative single case-study, which reveals specific developmental patterns in the context of the specific task. Implications and future work are discussed regarding the development of intelligent robotic systems that allow child-initiated interaction as well as targeted and not constant robot interventions.

This study examines the coiling and uncoiling motions of a soft pneumatic actuator inspired by the awn tissue of Erodium cicutarium. These tissues have embedded cellulose fibers distributed in a tilted helical pattern, which induces hygroscopic coiling and uncoiling in response to the daily changes in ambient humidity. Such sophisticated motions can eventually “drill” the seed at the tip of awn tissue into the soil: a drill bit in the plant kingdom. Through finite element simulation and experimental testing, this study examines a soft pneumatic actuator that has a similar reinforcing fiber layout to the Erodium plant tissue. This actuator, in essence, is a thin-walled elastomeric cylinder covered by tilted helical Kevlar fibers. Upon internal pressurization, it can exhibit a coiling motion by a combination of simultaneous twisting, bending, and extension. Parametric analyses show that the coiling motion characteristics are directly related to the geometry of tilted helical fibers. Notably, a moderate tilt in the reinforcing helical fiber leads to many coils of small radius, while a significant tilt gives fewer coils of larger radius. The results of this study can offer guidelines for constructing plant-inspired robotic manipulators that can achieve complicated motions with simple designs.

In recent years the field of soft robotics has gained a lot of interest both in academia and industry. In contrast to rigid robots, which are potentially very powerful and precise, soft robots are composed of compliant materials like gels or elastomers (Rich et al., 2018; Majidi, 2019). Their exclusive composition of nearly entirely soft materials offers the potential to extend the use of robotics to fields like healthcare (Burgner-Kahrs et al., 2015; Banerjee et al., 2018) and advance the emerging domain of cooperative human-machine interaction (Asbeck et al., 2014). One material class used frequently in soft robotics as actuators are electroactive polymers (EAPs). Especially dielectric elastomer actuators (DEAs) consisting of a thin elastomer membrane sandwiched between two compliant electrodes offer promising characteristics for actuator drives (Pelrine et al., 2000). Under an applied electric field, the resulting electrostatic pressure leads to a reduction in thickness and an expansion in the free spatial directions. The resulting expansion can reach strain levels of more than 300% (Bar-Cohen, 2004). This paper presents a bioinspired worm-like crawling robot based on DEAs with additional textile reinforcement in its silicone structures. A special focus is set on the developed cylindrical actuator segments that act as linear actuators.

This paper presents a three-layered hybrid collision avoidance (COLAV) system for autonomous surface vehicles, compliant with rules 8 and 13–17 of the International Regulations for Preventing Collisions at Sea (COLREGs). The COLAV system consists of a high-level planner producing an energy-optimized trajectory, a model-predictive-control-based mid-level COLAV algorithm considering moving obstacles and the COLREGs, and the branching-course model predictive control algorithm for short-term COLAV handling emergency situations in accordance with the COLREGs. Previously developed algorithms by the authors are used for the high-level planner and short-term COLAV, while we in this paper further develop the mid-level algorithm to make it comply with COLREGs rules 13–17. This includes developing a state machine for classifying obstacle vessels using a combination of the geometrical situation, the distance and time to the closest point of approach (CPA) and a new CPA-like measure. The performance of the hybrid COLAV system is tested through numerical simulations for three scenarios representing a range of different challenges, including multi-obstacle situations with multiple simultaneously active COLREGs rules, and also obstacles ignoring the COLREGs. The COLAV system avoids collision in all the scenarios, and follows the energy-optimized trajectory when the obstacles do not interfere with it.

While direct local communication is very important for the organization of robot swarms, so far it has mostly been used for relatively simple tasks such as signaling robots preferences or states. Inspired by the emergence of meaning found in natural languages, more complex communication skills could allow robot swarms to tackle novel situations in ways that may not be a priori obvious to the experimenter. This would pave the way for the design of robot swarms with higher autonomy and adaptivity. The state of the art regarding the emergence of communication for robot swarms has mostly focused on offline evolutionary approaches, which showed that signaling and communication can emerge spontaneously even when not explicitly promoted. However, these approaches do not lead to complex, language-like communication skills, and signals are tightly linked to environmental and/or sensory-motor states that are specific to the task for which communication was evolved. To move beyond current practice, we advocate an approach to emergent communication in robot swarms based on language games. Thanks to language games, previous studies showed that cultural self-organization—rather than biological evolution—can be responsible for the complexity and expressive power of language. We suggest that swarm robotics can be an ideal test-bed to advance research on the emergence of language-like communication. The latter can be key to provide robot swarms with additional skills to support self-organization and adaptivity, enabling the design of more complex collective behaviors.

Daily human activity is characterized by a broad variety of movement tasks. This work summarizes the sagittal hip, knee, and ankle joint biomechanics for a broad range of daily movements, based on previously published literature, to identify requirements for robotic design. Maximum joint power, moment, angular velocity, and angular acceleration, as well as the movement-related range of motion and the mean absolute power were extracted, compared, and analyzed for essential and sportive movement tasks. We found that the full human range of motion is required to mimic human like performance and versatility. In general, sportive movements were found to exhibit the highest joint requirements in angular velocity, angular acceleration, moment, power, and mean absolute power. However, at the hip, essential movements, such as recovery, had comparable or even higher requirements. Further, we found that the moment and power demands were generally higher in stance, while the angular velocity and angular acceleration were mostly higher or equal in swing compared to stance for locomotion tasks. The extracted requirements provide a novel comprehensive overview that can help with the dimensioning of actuators enabling tailored assistance or rehabilitation for wearable lower limb robots, and to achieve essential, sportive or augmented performances that exceed natural human capabilities with humanoid robots.

Telerobotics aims to transfer human manipulation skills and dexterity over an arbitrary distance and at an arbitrary scale to a remote workplace. A telerobotic system that is transparent enables a natural and intuitive interaction. We postulate that embodiment (with three sub-components: sense of ownership, agency, and self-location) of the robotic system leads to optimal perceptual transparency and increases task performance. However, this has not yet been investigated directly. We reason along four premises and present findings from the literature that substantiate each of them: (1) the brain can embody non-bodily objects (e.g., robotic hands), (2) embodiment can be elicited with mediated sensorimotor interaction, (3) embodiment is robust against inconsistencies between the robotic system and the operator's body, and (4) embodiment positively correlates to dexterous task performance. We use the predictive encoding theory as a framework to interpret and discuss the results reported in the literature. Numerous previous studies have shown that it is possible to induce embodiment over a wide range of virtual and real extracorporeal objects (including artificial limbs, avatars, and android robots) through mediated sensorimotor interaction. Also, embodiment can occur for non-human morphologies including for elongated arms and a tail. In accordance with the predictive encoding theory, none of the sensory modalities is critical in establishing ownership, and discrepancies in multisensory signals do not necessarily lead to loss of embodiment. However, large discrepancies in terms of multisensory synchrony or visual likeness can prohibit embodiment from occurring. The literature provides less extensive support for the link between embodiment and (dexterous) task performance. However, data gathered with prosthetic hands do indicate a positive correlation. We conclude that all four premises are supported by direct or indirect evidence in the literature, suggesting that embodiment of a remote manipulator may improve dexterous performance in telerobotics. This warrants further implementation testing of embodiment in telerobotics. We formulate a first set of guidelines to apply embodiment in telerobotics and identify some important research topics.

Dramatic cost savings, safety improvements and accelerated nuclear decommissioning are all possible through the application of robotic solutions. Remotely-controlled systems with modern sensing capabilities, actuators and cutting tools have the potential for use in extremely hazardous environments, but operation in facilities used for handling radioactive material presents complex challenges for electronic components. We present a methodology and results obtained from testing in a radiation cell in which we demonstrate the operation of a robotic arm controlled using modern electronics exposed at 10 Gy/h to simulate radioactive conditions in the most hazardous nuclear waste handling facilities.

As robots make their way out of factories into human environments, outer space, and beyond, they require the skill to manipulate their environment in multifarious, unforeseeable circumstances. With this regard, pushing is an essential motion primitive that dramatically extends a robot's manipulation repertoire. In this work, we review the robotic pushing literature. While focusing on work concerned with predicting the motion of pushed objects, we also cover relevant applications of pushing for planning and control. Beginning with analytical approaches, under which we also subsume physics engines, we then proceed to discuss work on learning models from data. In doing so, we dedicate a separate section to deep learning approaches which have seen a recent upsurge in the literature. Concluding remarks and further research perspectives are given at the end of the paper.

Underwater robots are nowadays employed for many different applications; during the last decades, a wide variety of robotic vehicles have been developed by both companies and research institutes, different in shape, size, navigation system, and payload. While the market needs to constitute the real benchmark for commercial vehicles, novel approaches developed during research projects represent the standard for academia and research bodies. An interesting opportunity for the performance comparison of autonomous vehicles lies in robotics competitions, which serve as an useful testbed for state-of-the-art underwater technologies and a chance for the constructive evaluation of strengths and weaknesses of the participating platforms. In this framework, over the last few years, the Department of Industrial Engineering of the University of Florence participated in multiple robotics competitions, employing different vehicles. In particular, in September 2017 the team from the University of Florence took part in the European Robotics League Emergency Robots competition held in Piombino (Italy) using FeelHippo AUV, a compact and lightweight Autonomous Underwater Vehicle (AUV). Despite its size, FeelHippo AUV possesses a complete navigation system, able to offer good navigation accuracy, and diverse payload acquisition and analysis capabilities. This paper reports the main field results obtained by the team during the competition, with the aim of showing how it is possible to achieve satisfying performance (in terms of both navigation precision and payload data acquisition and processing) even with small-size vehicles such as FeelHippo AUV.

The aim of this study was to assess what drives gender-based differences in the experience of cybersickness within virtual environments. In general, those who have studied cybersickness (i.e., motion sickness associated with virtual reality [VR] exposure), oftentimes report that females are more susceptible than males. As there are many individual factors that could contribute to gender differences, understanding the biggest drivers could help point to solutions. Two experiments were conducted in which males and females were exposed for 20 min to a virtual rollercoaster. In the first experiment, individual factors that may contribute to cybersickness were assessed via self-report, body measurements, and surveys. Cybersickness was measured via the simulator sickness questionnaire and physiological sensor data. Interpupillary distance (IPD) non-fit was found to be the primary driver of gender differences in cybersickness, with motion sickness susceptibility identified as a secondary driver. Females whose IPD could not be properly fit to the VR headset and had a high motion sickness history suffered the most cybersickness and did not fully recover within 1 h post exposure. A follow-on experiment demonstrated that when females could properly fit their IPD to the VR headset, they experienced cybersickness in a manner similar to males, with high cybersickness immediately upon cessation of VR exposure but recovery within 1 h post exposure. Taken together, the results suggest that gender differences in cybersickness may be largely contingent on whether or not the VR display can be fit to the IPD of the user; with a substantially greater proportion of females unable to achieve a good fit. VR displays may need to be redesigned to have a wider IPD adjustable range in order to reduce cybersickness rates, especially among females.

Robots face a rapidly expanding range of potential applications beyond controlled environments, from remote exploration and search-and-rescue to household assistance and agriculture. The focus of physical interaction is typically delegated to end-effectors—fixtures, grippers or hands—as these machines perform manual tasks. Yet, effective deployment of versatile robot hands in the real world is still limited to few examples, despite decades of dedicated research. In this paper we review hands that found application in the field, aiming to discuss open challenges with more articulated designs, discussing novel trends and perspectives. We hope to encourage swift development of capable robotic hands for long-term use in varied real world settings. The first part of the paper centers around progress in artificial hand design, identifying key functions for a variety of environments. The final part focuses on the overall trends in hand mechanics, sensors and control, and how performance and resiliency are qualified for real world deployment.

Path planning is general problem of mobile robots, which has special characteristics when applied to marine applications. In addition to avoid colliding with obstacles, in marine scenarios, environment conditions such as water currents or wind need to be taken into account in the path planning process. In this paper, several solutions based on the Fast Marching Method are proposed. The basic method focus on collision avoidance and optimal planning and, later on, using the same underlying method, the influence of marine currents in the optimal path planning is detailed. Finally, the application of these methods to consider marine robot formations is presented.

Emotional deception and emotional attachment are regarded as ethical concerns in human-robot interaction. Considering these concerns is essential, particularly as little is known about longitudinal effects of interactions with social robots. We ran a longitudinal user study with older adults in two retirement villages, where people interacted with a robot in a didactic setting for eight sessions over a period of 4 weeks. The robot would show either non-emotive or emotive behavior during these interactions in order to investigate emotional deception. Questionnaires were given to investigate participants' acceptance of the robot, perception of the social interactions with the robot and attachment to the robot. Results show that the robot's behavior did not seem to influence participants' acceptance of the robot, perception of the interaction or attachment to the robot. Time did not appear to influence participants' level of attachment to the robot, which ranged from low to medium. The perceived ease of using the robot significantly increased over time. These findings indicate that a robot showing emotions—and perhaps resulting in users being deceived—in a didactic setting may not by default negatively influence participants' acceptance and perception of the robot, and that older adults may not become distressed if the robot would break or be taken away from them, as attachment to the robot in this didactic setting was not high. However, more research is required as there may be other factors influencing these ethical concerns, and support through other measurements than questionnaires is required to be able to draw conclusions regarding these concerns.

Natural language is inherently a discrete symbolic representation of human knowledge. Recent advances in machine learning (ML) and in natural language processing (NLP) seem to contradict the above intuition: discrete symbols are fading away, erased by vectors or tensors called distributed and distributional representations. However, there is a strict link between distributed/distributional representations and discrete symbols, being the first an approximation of the second. A clearer understanding of the strict link between distributed/distributional representations and symbols may certainly lead to radically new deep learning networks. In this paper we make a survey that aims to renew the link between symbolic representations and distributed/distributional representations. This is the right time to revitalize the area of interpreting how discrete symbols are represented inside neural networks.

Most collaborative tasks require interaction with everyday objects (e.g., utensils while cooking). Thus, robots must perceive everyday objects in an effective and efficient way. This highlights the necessity of understanding environmental factors and their impact on visual perception, such as illumination changes throughout the day on robotic systems in the real world. In object recognition, two of these factors are changes due to illumination of the scene and differences in the sensors capturing it. In this paper, we will present data augmentations for object recognition that enhance a deep learning architecture. We will show how simple linear and non-linear illumination models and feature concatenation can be used to improve deep learning-based approaches. The aim of this work is to allow for more realistic Human-Robot Interaction scenarios with a small amount of training data in combination with incremental interactive object learning. This will benefit the interaction with the robot to maximize object learning for long-term and location-independent learning in unshaped environments. With our model-based analysis, we showed that changes in illumination affect recognition approaches that use Deep Convolutional Neural Network to encode features for object recognition. Using data augmentation, we were able to show that such a system can be modified toward a more robust recognition without retraining the network. Additionally, we have shown that using simple brightness change models can help to improve the recognition across all training set sizes.

Repertoire-based learning is a data-efficient adaptation approach based on a two-step process in which (1) a large and diverse set of policies is learned in simulation, and (2) a planning or learning algorithm chooses the most appropriate policies according to the current situation (e.g., a damaged robot, a new object, etc.). In this paper, we relax the assumption of previous works that a single repertoire is enough for adaptation. Instead, we generate repertoires for many different situations (e.g., with a missing leg, on different floors, etc.) and let our algorithm selects the most useful prior. Our main contribution is an algorithm, APROL (Adaptive Prior selection for Repertoire-based Online Learning) to plan the next action by incorporating these priors when the robot has no information about the current situation. We evaluate APROL on two simulated tasks: (1) pushing unknown objects of various shapes and sizes with a robotic arm and (2) a goal reaching task with a damaged hexapod robot. We compare with “Reset-free Trial and Error” (RTE) and various single repertoire-based baselines. The results show that APROL solves both the tasks in less interaction time than the baselines. Additionally, we demonstrate APROL on a real, damaged hexapod that quickly learns to pick compensatory policies to reach a goal by avoiding obstacles in the path.

Automation of logistic tasks, such as object picking and placing, is currently one of the most active areas of research in robotics. Handling delicate objects, such as fruits and vegetables, both in warehouses and in plantations, is a big challenge due to the delicacy and precision required for the task. This paper presents the CLASH hand, a Compliant Low-Cost Antagonistic Servo Hand, whose kinematics was specifically designed for handling groceries. The main feature of the hand is its variable stiffness, which allows it to withstand collisions with the environment and also to adapt the passive stiffness to the object weight while relying on a modular design using off-the-shelf low-cost components. Due to the implementation of differentially coupled flexors, the hand can be actuated like an underactuated hand but can also be driven with different stiffness levels to planned grasp poses, i.e., it can serve for both model-based grasp planning and for underactuated or model-free grasping. The hand also includes self-checking and logging processes, which enable more robust performance during grasping actions. This paper presents key aspects of the hand design, examines the robustness of the hand in impact tests, and uses a standardized fruit benchmarking test to verify the behavior of the hand when different actuator and sensor failures occur and are compensated for autonomously by the hand.

In the immediate aftermath following a large-scale release of radioactive material into the environment, it is necessary to determine the spatial distribution of radioactivity quickly. At present, this is conducted by utilizing manned aircraft equipped with large-volume radiation detection systems. Whilst these are capable of mapping large areas quickly, they suffer from a low spatial resolution due to the operating altitude of the aircraft. They are also expensive to deploy and their manned nature means that the operators are still at risk of exposure to potentially harmful ionizing radiation. Previous studies have identified the feasibility of utilizing unmanned aerial systems (UASs) in monitoring radiation in post-disaster environments. However, the majority of these systems suffer from a limited range or are too heavy to be easily integrated into regulatory restrictions that exist on the deployment of UASs worldwide. This study presents a new radiation mapping UAS based on a lightweight (8 kg) fixed-wing unmanned aircraft and tests its suitability to mapping post-disaster radiation in the Chornobyl Exclusion Zone (CEZ). The system is capable of continuous flight for more than 1 h and can resolve small scale changes in dose-rate in high resolution (sub-20 m). It is envisaged that with some minor development, these systems could be utilized to map large areas of hazardous land without exposing a single operator to a harmful dose of ionizing radiation.