Frontiers in Robotics and AI最新文献

The rapidly increasing capabilities of autonomous mobile robots promise to make them ubiquitous in the coming decade. These robots will continue to enhance efficiency and safety in novel applications such as disaster management, environmental monitoring, bridge inspection, and agricultural inspection. To operate autonomously without constant human intervention, even in remote or hazardous areas, robots must sense, process, and interpret environmental data using only onboard sensing and computation. This capability is made possible by advancements in perception algorithms, allowing these robots to rely primarily on their perception capabilities for navigation tasks. However, tiny robot autonomy is hindered mainly by sensors, memory, and computing due to size, area, weight, and power constraints. The bottleneck in these robots lies in the real-time perception in resource-constrained robots. To enable autonomy in robots of sizes that are less than 100 mm in body length, we draw inspiration from tiny organisms such as insects and hummingbirds, known for their sophisticated perception, navigation, and survival abilities despite their minimal sensor and neural system. This work aims to provide insights into designing a compact and efficient minimal perception framework for tiny autonomous robots from higher cognitive to lower sensor levels.

We live in a visual world where text cues are abundant in urban environments. The premise for our work is for robots to capitalize on these text features for visual place recognition. A new technique is introduced that uses an end-to-end scene text detection and recognition technique to improve robot localization and mapping through Visual Place Recognition (VPR). This technique addresses several challenges such as arbitrary shaped text, illumination variation, and occlusion. The proposed model captures text strings and associated bounding boxes specifically designed for VPR tasks. The primary contribution of this work is the utilization of an end-to-end scene text spotting framework that can effectively capture irregular and occluded text in diverse environments. We conduct experimental evaluations on the Self-Collected TextPlace (SCTP) benchmark dataset, and our approach outperforms state-of-the-art methods in terms of precision and recall, which validates the effectiveness and potential of our proposed approach for VPR.

In recent years, the development of robots for agro-industrial applications, such as the cultivation of Solanum tuberosum potatoes, has aroused the interest of the academic and scientific communities. This is due, at least in part, to the complexity of modeling and robustly controlling some dynamics inherent to nonlinear behaviors normally attributed to the different technologies associated with the movement of these autonomous vehicles and their non-holonomic constraints. The different nonlinear dynamics of mobile robots are usually represented by state-space models. However, given some equilibrium and stability characteristics, the implementation of effective controllers for the robust parametric tracking and variation problem requires techniques that allow the operability of robots around regions of stable equilibrium. Feedback linearization control is one such technique that attempts to mathematically eliminate nonlinear expressions from the plant model. However, this technique requires an observable and controllable mathematical model. If there is some relationship between the model inputs and a controlled output that allows the relative degree of the control law to be determined, the controller design and implementation are posed as a linear issue. Flat filters developed from the generalized proportional integral control approach are an alternative that could facilitate the design of controllers for these linearized systems. From these flat filters, it is possible to obtain the transfer function of a controller without relying on the derivatives of the system output. This work proposes the design of a controller via exact linearization and its equivalent flat filter for a robot inspector of the soil resource of S. tuberosum crops in the department of Cundinamarca, Colombia. The actuator motion constraints resulted in a robot with two degrees of mobility and one non-holonomic constraint. Numerical validation of this system suggests that it can be an effective solution to the problem of tracking control at changing references by providing a system capable of navigating through crop rows. The results suggest correct tracking for linear and circular trajectories. However, the control lacks the ability to track spiral-type trajectories.

Diabetic Retinopathy (DR) is a serious eye condition that occurs due to high blood sugar levels in patients with Diabetes Mellitus. If left untreated, DR can potentially result in blindness. Using automated neural network-based methods to grade DR shows potential for early detection. However, the uneven and non-quadrilateral forms of DR lesions provide difficulties for traditional Convolutional Neural Network (CNN)-based architectures. To address this challenge and explore a novel algorithm architecture, this work delves into the usage of contrasting cluster assignments in retinal fundus images with the Swapping Assignments between multiple Views (SwAV) algorithm for DR grading. An ablation study was made where SwAV outperformed other CNN and Transformer-based models, independently and in ensemble configurations with an accuracy of 87.00% despite having fewer parameters and layers. The proposed approach outperforms existing state-of-the-art models regarding classification metrics, complexity, and prediction time. The findings offer great potential for medical practitioners, allowing for more accurate diagnosis of DR and earlier treatments to avoid visual loss.

Interaction is a dynamic process that evolves in real time. Participants interpret and orient themselves towards turns of speech based on expectations of relevance and social/conversational norms (that have been extensively studied in the field of Conversation analysis). A true challenge to Human Robot Interaction (HRI) is to develop a system capable of understanding and adapting to the changing context, where the meaning of a turn is construed based on the turns that have come before. In this work, we identify issues arising from the inadequate handling of the sequential flow within a corpus of in-the-wild HRIs in an open-world university library setting. The insights gained from this analysis can be used to guide the design of better systems capable of handling complex situations. We finish by surveying efforts to mitigate the identified problems from a natural language processing/machine dialogue management perspective.

Introduction: Slip-related falls are a significant concern, particularly for vulnerable populations such as the elderly and individuals with gait disorders, necessitating effective preventive measures. This manuscript presents a biomechanical study of how the lower limbs react to perturbations that can trigger a slip-like fall, with the ultimate goal of identifying target specifications for developing a wearable robotic system for slip-like fall prevention.

Methods: Our analysis provides a comprehensive understanding of the natural human biomechanical response to slip perturbations in both slipping and trailing legs, by innovatively collecting parameters from both the sagittal and frontal plane since both play pivotal roles in maintaining stability and preventing falls and thus provide new insights to fall prevention. We investigated various external factors, including gait speed, surface inclination, slipping foot, and perturbation intensity, while collecting diverse data sets encompassing kinematic, spatiotemporal parameters, electromyographic data, as well as torque, range of motion, rotations per minute, detection, and actuation times.

Results: The biomechanical response to slip-like perturbations by the hips, knees, and ankles of the slipping leg was characterized by extension, flexion, and plantarflexion moments, respectively. In the trailing leg, responses included hip flexion, knee extension, and ankle plantarflexion. Additionally, these responses were influenced by gait speed, surface inclination, and perturbation intensity. Our study identified target range of motion parameters of 85.19°, 106.34°, and 95.23° for the hips, knees, and ankles, respectively. Furthermore, rotations per minute values ranged from 17.85 to 51.10 for the hip, 21.73 to 63.80 for the knee, and 17.52 to 57.14 for the ankle joints. Finally, flexion/extension torque values were estimated as -3.05 to 3.22 Nm/kg for the hip, -1.70 to 2.34 Nm/kg for the knee, and -2.21 to 0.90 Nm/kg for the ankle joints.

Discussion: This study contributes valuable insights into the biomechanical aspects of slip-like fall prevention and informs the development of wearable robotic systems to enhance safety in vulnerable populations.

Introduction: Soft pneumatic actuators (SPAs) play a pivotal role in soft robotics due to their unique characteristics of compliance, flexibility, and adaptability. There are plenty of approaches that examine the modeling parameters of SPAs, aiming to optimize their design and, thus, achieve the most advantageous responses. Current optimization methods applied to SPAs are usually performed individually for each design parameter without considering the simultaneous effect all parameters can have on the output performance. This modeling shortcoming is essential to be addressed since customized SPAs are used in a variety of applications, each with different output requirements.

Methods: This study provides a generalized design optimization framework for modeling the SPA performance for their motion profiles, the produced strain energy while being deformed, and their stiffness characteristics. Utilizing experimentally validated finite element methods, all geometrical and material parameters of the models are investigated in response surface methodology optimization using the central composite design approach.

Results: The results showcase the entire design space of omnidirectional SPAs, along with their output performance, providing guidelines to the end user for design optimization.

Discussion: The offering of this modeling process for SPAs can be adapted to the demands of any potential application and ensure the best performance with respect to the required output responses.

Introduction: Robots are being introduced into increasingly social environments. As these robots become more ingrained in social spaces, they will have to abide by the social norms that guide human interactions. At times, however, robots will violate norms and perhaps even deceive their human interaction partners. This study provides some of the first evidence for how people perceive and evaluate robot deception, especially three types of deception behaviors theorized in the technology ethics literature: External state deception (cues that intentionally misrepresent or omit details from the external world: e.g., lying), Hidden state deception (cues designed to conceal or obscure the presence of a capacity or internal state the robot possesses), and Superficial state deception (cues that suggest a robot has some capacity or internal state that it lacks).

Methods: Participants (N = 498) were assigned to read one of three vignettes, each corresponding to one of the deceptive behavior types. Participants provided responses to qualitative and quantitative measures, which examined to what degree people approved of the behaviors, perceived them to be deceptive, found them to be justified, and believed that other agents were involved in the robots' deceptive behavior.

Results: Participants rated hidden state deception as the most deceptive and approved of it the least among the three deception types. They considered external state and superficial state deception behaviors to be comparably deceptive; but while external state deception was generally approved, superficial state deception was not. Participants in the hidden state condition often implicated agents other than the robot in the deception.

Conclusion: This study provides some of the first evidence for how people perceive and evaluate the deceptiveness of robot deception behavior types. This study found that people people distinguish among the three types of deception behaviors and see them as differently deceptive and approve of them differently. They also see at least the hidden state deception as stemming more from the designers than the robot itself.

This paper introduces software patterns (registration, acquire-release, and cache awareness) and data structures (Petri net, finite state machine, and protocol flag array) to support the coordinated execution of software activities (also called "components" or "agents"). Moreover, it presents and tests an implementation for Petri nets that supports real-time execution in shared memory for deployment inside one individual robot and separates event firing and handling, enabling distributed deployment between multiple robots. Experimental validation of the introduced patterns and data structures is performed within the context of activities for task execution, control and perception, and decision making for an application on coordinated navigation.