Frontiers in Robotics and AI最新文献

The importance of simulating patient behavior for medical assessment training has grown in recent decades due to the increasing variety of simulation tools, including standardized/simulated patients, humanoid and android robot-patients. Yet, there is still a need for improvement of current android robot-patients to accurately simulate patient behavior, among which taking into account their hearing loss is of particular importance. This paper is the first to consider hearing loss simulation in an android robot-patient and its results provide valuable insights for future developments. For this purpose, an open-source dataset of audio data and audiograms from human listeners was used to simulate the effect of hearing loss on an automatic speech recognition (ASR) system. The performance of the system was evaluated in terms of both word error rate (WER) and word information preserved (WIP). Comparing different ASR models commonly used in robotics, it appears that the model size alone is insufficient to predict ASR performance in presence of simulated hearing loss. However, though absolute values of WER and WIP do not predict the intelligibility for human listeners, they do highly correlate with it and thus could be used, for example, to compare the performance of hearing aid algorithms.

This paper presents a tangle- and contact-free path planning (TCFPP) for a mobile robot attached to a base station with a finite-length cable. This type of robot, called a tethered mobile robot, can endure long-time exploration with a continuous power supply and stable communication via its cable. However, the robot faces potential hazards that endanger its operation such as cable snagging on and cable entanglement with obstacles and the robot. To address these challenges, our approach incorporates homotopy-aware path planning into deep reinforcement learning. The proposed reward design in the learning problem penalizes the cable-obstacle and cable-robot contacts and encourages the robot to follow the homotopy-aware path toward a goal. We consider two distinct scenarios for the initial cable configuration: 1) the robot pulls the cable sequentially from the base while heading for the goal, and 2) the robot moves to the goal starting from a state where the cable has already been partially deployed. The proposed method is compared with naive approaches in terms of contact avoidance and path similarity. Simulation results revealed that the robot can successfully find a contact-minimized path under the guidance of the reference path in both scenarios.

Much of the Earth and many surfaces of extraterrestrial bodies are composed of non-cohesive particulate matter. Locomoting on such granular terrain is challenging for common robotic devices, either wheeled or legged. In this work, we discover a robust alternative locomotion mechanism on granular media-generating movement via self-vibration. To demonstrate the effectiveness of this locomotion mechanism, we develop a cube-shaped robot with an embedded vibratory motor and conduct systematic experiments on granular terrains of various particle properties and slopes. We investigate how locomotion changes as a function of vibration frequency/intensity on such granular terrains. Compared to hard surfaces, we find such a vibratory locomotion mechanism enables the robot to move faster, and more stably on granular surfaces, facilitated by the interaction between the body and surrounding grains. We develop a numerical simulation of a vibrating single cube on granular media, enabling us to justify our hypothesis that the cube achieves locomotion through the oscillations excited at a distance from the cube's center of mass. The simplicity in structural design and controls of this robotic system indicates that vibratory locomotion can be a valuable alternative way to produce robust locomotion on granular terrains. We further demonstrate that such cube-shaped robots can be used as modular units for vibratory robots with capabilities of maneuverable forward and turning motions, showing potential practical scenarios for robotic systems.

Transforming planar structures into volumetric objects typically requires manual folding processes, akin to origami. However, manual intervention at sub-centimeter scales is impractical. Instead, folding is achieved using volume-changing smart materials that respond to physical or chemical stimuli, be it with direct contact such as hydration, pH, or remotely e.g., light or magnetism. The complexity of small-scale structures often restricts the variety of smart materials used and the number of folding sequences. In this study, we propose a method to sequentially self-fold millimeter scale origami using magnetic induction heating at $150$ kHz and 3.2 mT. Additionally, we introduce a method for designing self-folding overhand knots and predicting the folding sequence using the magneto-thermal model we developed. This methodology is demonstrated to sequentially self-fold by optimizing the surface, placement, and geometry of metal workpieces, and is validated through the self-folding of various structures, including a 380 ${mm}^2$ croissant, a $321$ mm² box, a $447$ mm² bio-mimetic Mimosa pudica leaf, and an overhand knot covering $524$ mm². Our work shows significant potential for miniature self-folding origami robots owing to the novel sequential folding approach and the ability to achieve remote and tetherless self-folding within constrained environments.

This paper introduces a compact end-effector ankle rehabilitation robot (CEARR) system for addressing ankle range of motion (ROM) rehabilitation. The CEARR features a bilaterally symmetrical rehabilitation structure, with each side possessing three degrees of freedom (DOF) driven by three independently designed actuators. The working intervals of each actuator are separated by a series connection, ensuring they operate without interference to accommodate the dorsiflexion/plantarflexion (DO/PL), inversion/eversion (IN/EV), and adduction/abduction (AD/AB) DOF requirements for comprehensive ankle rehabilitation. In addition, we integrated an actuator and foldable brackets to accommodate patients in varied postures. We decoded the motor intention based on the surface electromyography (sEMG) and torque signals generated by the subjects' ankle joints in voluntary rehabilitation. Besides, we designed a real-time voluntary-triggered control (VTC) strategy to enhance the rehabilitation effect, in which the root mean square (RMS) of sEMG was utilized to trigger and adjust the CEARR rehabilitation velocity support. We verified the consistency of voluntary movement with CEARR rehabilitation support output for four healthy subjects on a nonlinear sEMG signal with an $R^2$ metric of approximately 0.67. We tested the consistency of triggering velocity trends with a linear torque signal for one healthy individual with an $R^2$ metric of approximately 0.99.

When designing social robots for educational settings, there is often an emphasis on domain knowledge. This presents challenges: 1) Either robots must autonomously acquire domain knowledge, a currently unsolved problem in HRI, or 2) the designers provide this knowledge implying re-programming the robot for new contexts. Recent research explores alternative, relatively easier to port, knowledge areas like student rapport, engagement, and synchrony though these constructs are typically treated as the ultimate goals, when the final goal should be students' learning. Our aim is to propose a shift in how engagement is considered, aligning it naturally with learning. We introduce the notion of a skilled ignorant peer robot: a robot peer that has little to no domain knowledge but possesses knowledge of student behaviours conducive to learning, i.e., behaviours indicative of productive engagement as extracted from student behavioral profiles. We formally investigate how such a robot's interventions manipulate the children's engagement conducive to learning. Specifically, we evaluate two versions of the proposed robot, namely, Harry and Hermione, in a user study with 136 students where each version differs in terms of the intervention strategy. Harry focuses on which suggestions to intervene with from a pool of communication, exploration, and reflection inducing suggestions, while Hermione also carefully considers when and why to intervene. While the teams interacting with Harry have higher productive engagement correlated to learning, this engagement is not affected by the robot's intervention scheme. In contrast, Hermione's well-timed interventions, deemed more useful, correlate with productive engagement though engagement is not correlated to learning. These results highlight the potential of a social educational robot as a skilled ignorant peer and stress the importance of precisely timing the robot interventions in a learning environment to be able to manipulate moderating variable of interest such as productive engagement.

Introduction: Paediatric forearm fractures are a prevalent reason for medical consultation, often requiring diagnostic X-rays that present a risk due to ionising radiation, especially concerning given the sensitivity of children's tissues. This paper explores the efficacy of ultrasound imaging, particularly through the development of the SonoBox system, as a safer, non-ionising alternative. With emerging evidence supporting ultrasound as a viable method for fracture assessment, innovations like SonoBox will become increasingly important.

Materials and methods: In our project, we want to advance ultrasound-based, contact-free, and automated cross-sectional imaging for diagnosing paediatric forearm fractures. To this end, we are building a technical platform that navigates a commercially available ultrasound probe around the extremity within a water-filled tank, utilising intelligent robot control and image processing methods to generate a comprehensive ultrasound tomogram. Safety and hygiene considerations, gender and diversity relevance, and the potential reduction of radiation exposure and examination pain are pivotal aspects of this endeavour.

Results: Preliminary experiments have demonstrated the feasibility of rapidly generating ultrasound tomographies in a water bath, overcoming challenges such as water turbulence during probe movement. The SonoBox prototype has shown promising results in transmitting position data for ultrasound imaging, indicating potential for autonomous, accurate, and potentially painless fracture diagnosis. The project outlines further goals, including the construction of prototypes, validation through patient studies, and development of a hygiene concept for clinical application.

Conclusion: The SonoBox project represents a significant step forward in paediatric fracture diagnostics, offering a safer, more comfortable alternative to traditional X-ray imaging. By automating the imaging process and removing the need for direct contact, SonoBox has the potential to improve clinical efficiency, reduce patient discomfort, and broaden the scope of ultrasound applications. Further research and development will focus on validating its effectiveness in clinical settings and exploring its utility in other medical and veterinary applications.

Physical interaction with patients, for example conducted as part of a diagnostic examination or surgical procedure, provides clinicians with a wealth of information about their condition. Simulating this interaction is of great interest to researchers in both haptics and medical education, and the development of softness changing tactile interfaces is important in recreating the feel of different soft tissues. This paper presents designs for a variety of novel electromechanical and electromagnetic mechanisms for controlling particle jamming-based, hardness changing tactile displays, intended to allow medical trainees to experience these physical interactions in a range of simulation settings such as clinical skills teaching laboratories. Each design is then subjected to a battery of mechanical tests to evaluate its effectiveness compared to the state of the art, as well as their suitability for simulating the physical hardness of different types of soft tissues, previously characterised in established literature. These results demonstrate that all of the technologies presented are able to exhibit a measurable hardness change, with Shore hardness values between 3A and 57A achieved by the most effective constriction-based device. The electromechanical devices based on constriction and compression, and the state-of-the-art pneumatic device, were able to achieve hardness changes within a range that is useful for replicating the softness of organic tissue. The electromechanical and electromagnetic devices were also found to effect their full range of hardness change in less than a second, compared to several seconds for the state-of-the-art. These results show that the performance of softness changing tactile displays can be improved with the electromechanical actuation techniques proposed in this paper, and that such displays are able to replicate the physical characteristics of soft tissues and may therefore be of benefit in medical training and simulation scenarios.

For a drive unit for axes of robots and robot-like systems (RLS) usually a motor-gearbox arrangement is chosen due to its high-power density. The combination of a high-ratio gearbox and a high-speed electric motor ensures a very compact and efficient design of the drive train. The transmission properties primarily determine the properties of the axes and the whole robot system. Robots and RLS use various types of high-ratio precision gearboxes based on different operating principles. Due to the different operating principles, it is difficult to describe comparable properties across all different types. In addition, there are many influences on the properties which significantly determine their shapes and values. These influencing parameters are insufficiently documented and are often poorly accessible for profound comparability and further consideration. In this paper, an overview of the properties of robot gearboxes is given. Based on these properties, different robot gearboxes can be systematically evaluated and compared to one another. The properties are influenced by various design, operating or manufacturing factors such as the gearbox size, the operating torque and speed or the manufacturing process. In a further step, these influences on the most relevant properties, efficiency and stiffness, are determined and systematically evaluated. This evaluation is based on the specification data of various robot gearbox manufacturers. The properties efficiency and stiffness show a dependency on the gearbox size, the operating torque, speed as well as the ambient temperature and on the transmission ratio. The shown procedure can also be adapted to other properties.

Introduction: This paper addresses the growing healthcare needs of an ageing population and the lack of advanced technologies with social capabilities that are cost effective, user friendly, and readily adopted. In response to this motivation, a socially assistive walker is designed to provide physical and cognitive support in activities of daily living for older adults.

Methods: Physical and cognitive support is provided by walker's structure, sensors, and feedback interfaces to assist users daily living activities, as well as, in navigating environment safely and efficiently. The walker's design arises from semi-structured interviews conducted with ageing experts, leading to the development of two levels or modes of social interaction, namely low and high interaction.

Results and discussion: In a cohort of 14 adults, the study found the device easy to use regardless of the interaction mode, with 78.5% expressing a preference for the version featuring embodiment, verbal feedback, and more proactive cues (p $<$ 0.05). The results also prompted ideas and suggestions for new designs based on insights gleaned from the user. This research contributes to the field of socially assistive robotics by offering an example of a user centred approach to address the healthcare challenges an ageing population poses.