Frontiers in Robotics and AI最新文献

Substation robots face significant challenges in path planning due to the complex electromagnetic environment, dense equipment layout, and safety-critical operational requirements. This paper proposes a path planning algorithm based on deep reinforcement learning enhanced by ant colony optimization, establishing a synergistic optimization framework that combines bio-inspired algorithms with deep learning. The proposed method addresses critical path planning issues in substation inspection and maintenance operations. The approach includes: 1) designing a pheromone-guided exploration strategy that transforms environmental prior knowledge into spatial bias to reduce ineffective exploration; 2) establishing a high-quality sample screening mechanism that enhances Q-network training through ant colony path experience to improve sample efficiency; 3) implementing dynamic decision weight adjustment that enables gradual transition from heuristic guidance to autonomous learning decisions. Experimental results in complex environments demonstrate the method's superiority. Compared to state-of-the-art baselines including PPO, DDQN, and A*, the proposed method achieves 24% higher sample efficiency, 18% reduction in average path length, and superior dynamic obstacle avoidance. Field validation in a 2,500-square-meter substation confirms a 14.8% improvement in task completion rate compared to standard DRL approaches.

Training mobile robots through digital twins with deep reinforcement learning (DRL) has gained increasing attention to ensure efficient and safe navigation in complex environments. In this paper, we propose a novel physics-inspired DRL framework that achieves both effective and explainable motion planning. We represent the robot, destination, and obstacles as electrical charges and model their interactions using Coulomb forces. These forces are incorporated into the reward function, providing both attractive and repulsive signals to guide robot behavior. In addition, obstacle boundaries extracted from LiDAR segmentation are integrated as anticipatory rewards, allowing the robot to avoid collisions from a distance. The proposed model is first trained in Gazebo simulation environments and subsequently deployed on a real TurtleBot v3 robot. Extensive experiments in both simulation and real-world scenarios demonstrate the effectiveness of the proposed framework. Results show that our method significantly reduces collisions, maintains safe distances from obstacles, and generates safer trajectories toward the destinations.

The proliferation of the Fourth Industrial Revolution (4IR) is transforming the accounting landscape, with technologies such as Robotic Process Automation (RPA) changing the face of traditional accounting processes. This study investigates the level of RPA adoption among accountants in South Africa and examines how technological-organizational-environmental (TOE) factors influence the behavioral intention of RPA adoption. The study employed an exploratory cross-sectional survey comprising responses from 100 professional accountants in practice to analyze its data, combining descriptive statistics with a multiple linear regression model supported by correlation tests to determine significant predictors of RPA adoption intention. The robustness of the model, which was verified by multiple pre- and post-analysis checks, indicated that institutional support, particularly normative pressure, has the strongest influence on adoption intention, with an adjusted R² value of 0.27 highly significant. This highlights the crucial role that organizational readiness, managerial support, and technology readiness play in enabling RPA adoption. On the other hand, mimetic pressure showed a negative influence, indicating that the industry-wide adoption of RPA technology may raise concerns and anxiety about job displacement. Overall, the findings reinforce the importance of organizational capacity-building in fostering RPA adoption while also revealing the complexity of environmental and technological factors that influence the adoption decisions of professional accountants in a developing-economy context. The findings support SDG 9 by emphasizing capacity building and inclusive digital transformation.

Introduction: Informal dementia caregivers face significant emotional and physical burdens, yet evidence-based interventions like REACH are often limited by high labor costs and scalability constraints.

Methods: We design a Robot-based Information and Support to Enhance Alzheimer's Caregiver Health (RISE) system, which uses novel social robotics and generative AI to deliver automated and personalized caregiver training and stress management. RISE uses retrieval-augmented generative AI (RAG-AI) grounded in the verified REACH Caregiver Notebook to ensure content safety and minimize hallucinations. It employs the social robot Pepper to deliver interactive presentations, Q&A sessions, review quizzes, and stress reduction activities. A technical evaluation and a two-phase user evaluation was conducted.

Results: We found that the RISE's RAG-AI backend achieved 87% correctness and 92% relevancy when compared to ground truth. User feedback indicated strong acceptance, with Likert-scale usability scores ranging from 3.6 to 4.6 out of 5 across all components.

Discussion: These results suggest that combining verifiable AI architectures with embodied social robotics offers a feasible, scalable solution for enhancing caregiver support and wellbeing. Future work could include a larger scale user study involving real informal dementia caregivers.

Introduction: This paper presents a robotic transesophageal echocardiography (TEE) system that replicates all essential degrees of freedom available in manual TEE procedures. The developed robotic system advances dual-subsystem architectures through enhanced mechanical design and deep learning-based kinematic modeling.

Methods: Building upon previous designs that manipulate the TEE probe from both handle and gastroscope tube, our system integrates with a teleoperated UR5 manipulator to accommodate both supine and left lateral decubitus patient positions, addressing the full spectrum of clinical TEE procedures. The system features 6 DOF at the probe handle and 2 DOF at the gastroscope tube. Together, these create optimal gastroscope tube geometry, minimizing cable tension asymmetry and friction-induced nonlinearities inherent in cable-driven mechanisms. The primary contribution is a data-driven kinematic model using recurrent neural networks with LSTM units that overcomes fundamental limitations of analytical approaches for continuum manipulators. Trained on 42,000 synchronized pose-command pairs collected across three gastroscope tube configurations (0°, 45°, 90° bends), the model effectively captures dead zones, hysteresis, and coupling effects between steering mechanisms.

Results: Experimental validation demonstrates strong position tracking across all three gastroscope tube configurations. The model achieves RMSE of 1.267 mm for the 0° configuration, 1.209 mm for the 45° configuration, and 1.194 mm for the 90° configuration. Mean orientation errors are 7.064° at 0°, 8.503° at 45°, and 4.947° at the clinically critical 90° configuration. The model exhibits coordinate frame independence with only 0.06 mm RMSE difference between original and rotated datasets. This confirms true kinematic learning rather than coordinate-specific patterns. With 1.8 ms inference time, the system achieves real-time performance essential for clinical deployment.

Discussion: This integration of robotic system design with deep learning establishes a foundation for semi-autonomous TEE systems. The developed system can support both diagnostic TEE examinations and TEE-guided structural heart interventions.

The subject of this article is the development of an unmanned surface vehicle (USV) for the removal of floating debris. A twin-hulled boat with four thrusters placed at the corners of the vessel is used for this purpose. The trash is collected in a storage space through a timing belt driven by an electric motor. The debris is accumulated in a funnel positioned at the front of the boat and subsequently raised through this belt into the garbage bin. The boat is equipped with a spherical camera, a long-range 2D LiDAR, and an inertial measurement unit (IMU) for simultaneous localization and mapping (SLAM). The floating debris is identified from rectified camera frames using YOLO, while the LiDAR and IMU concurrently provide the USV's odometry. Visual methods are utilized to determine the location of debris and obstacles in the 3D environment. The optimal order in which the debris is collected is determined by solving the orienteering problem, and the planar convex hull of the boat is combined with map and obstacle data via the Open Motion Planning Library (OMPL) to perform path planning. Pure pursuit is used to generate the trajectory from the obtained path. Limits on the linear and angular velocities are experimentally estimated, and a PID controller is tuned to improve path following. The USV is evaluated in an indoor swimming pool containing static obstacles and floating debris.

Tendon-driven continuum actuators (TDCAs) provide compliant and lifelike motion that is well suited for human-robot interaction, but their structural compliance and underactuation make them susceptible to undesired vibrations, particularly along unactuated axes under load. This work addresses vibration suppression in such systems by proposing a real-time control strategy for a two-degree-of-freedom TDCA-based soft robotic neck used in the HARU social robot, where yaw motion is unactuated and prone to oscillations due to eccentric loading. The proposed approach combines a current-based tendon pretensioning routine, baseline PID control of the actuated pitch and roll axes, and a novel Coupled Axis Indirect Vibration Suppression (CIVS) mechanism. CIVS exploits mechanical cross-axis coupling by using high-pass filtered yaw acceleration from an inertial sensor to generate transient tension modulations in the actuated tendons, thereby increasing effective damping of the unactuated yaw mode without introducing additional hardware or compromising compliance. A classical sliding mode control is also implemented as a nonlinear benchmark under identical hardware constraints. Experimental validation on the HARU neck under representative loading conditions demonstrates that the proposed method achieves substantial vibration attenuation. Compared to the baseline controller, CIVS reduces yaw angular range by approximately 53% and yaw acceleration area by over 60%, while preserving smooth, expressive motion. The results further show that CIVS outperforms the sliding mode controller in suppressing vibrations on the unactuated axis. These findings indicate that indirect, feedback-driven tendon modulation provides an effective and low-complexity solution for mitigating load-induced vibrations in underactuated soft robotic systems, making the approach particularly suitable for interactive applications where safety, compliance, and motion expressivity are critical.

Factor graph optimization serves as a fundamental framework for robotic perception, enabling applications such as pose estimation, simultaneous localization and mapping (SLAM), structure-from-motion (SfM), and situational modeling. Traditionally, these methods solve unconstrained least squares problems using algorithms such as Gauss-Newton and Levenberg-Marquardt. However, extending factor graphs with native support for hard equality constraints can yield more accurate state estimates and broaden their applicability, particularly in planning and control. Prior work has addressed equality handling either by soft penalties (large weights) or by nested-loop Augmented Lagrangian (AL) schemes. In this paper, we propose a novel extension of factor graphs that seamlessly incorporates hard equality constraints without requiring additional optimization techniques. Our approach maintains the efficiency and flexibility of existing second-order optimization techniques while ensuring constraint satisfaction. To validate the proposed method, an autonomous-vehicle velocity-tracking optimal control problem is solved and benchmarked against an AL baseline, both implemented in g2o. Additional comparisons are conducted in GTSAM, where the penalty method and AL are evaluated against our g2o implementations. Moreover, we introduce ecg2o, a header-only C++ library that extends the widely used g2o library with full support for hard equality-constrained optimization. This library, along with demonstrative examples and the optimal control problem, is available as open source at https://github.com/snt-arg/ecg2o.

Gaze is a fundamental aspect of non-verbal communication in human interaction, playing an important role in conveying attention, intentions, and emotions. A key concept in gaze-based human interaction is joint attention, the focus of two individuals on an object in a shared environment. In the context of human-robot interaction (HRI), gaze-following has become a growing research area, as it enables robots to appear more socially intelligent, engaging, and likable. While various technical approaches have been developed to achieve this capability, a comprehensive overview of existing implementations has been lacking. This scoping review addresses this gap by systematically categorizing existing solutions, offering a structured perspective on how gaze-following behavior is technically realized in the field of HRI. A systematic search was conducted across four databases, leading to the identification of 28 studies. To structure the findings, a taxonomy was developed that categorizes technological approaches along three key functional dimensions: (1) environment tracking, which involves recognizing the objects in the robot's surroundings; (2) gaze tracking, which refers to detecting and interpreting human gaze direction; and (3) gaze-environment mapping, which connects gaze information with objects in the shared environment to enable appropriate robotic responses. Across studies, a distinction emerges between constrained and unconstrained solutions. While constrained approaches, such as predefined object positions, provide high accuracy, they are often limited to controlled settings. In contrast, unconstrained methods offer greater flexibility but pose significant technical challenges. The complexity of the implementations also varies significantly, from simple rule-based approaches to advanced, adaptive systems that integrate multiple data sources. These findings highlight ongoing challenges in achieving robust and real-time gaze-following in robots, particularly in dynamic, real-world environments. Future research should focus on refining unconstrained tracking methods and leveraging advances in machine learning and computer vision to make human-robot interactions more natural and socially intuitive.

The ability to track the 6D pose distribution of an object while a mobile manipulator is still approaching it can enable the robot to pre-plan grasps, thereby improving both the time efficiency and robustness of mobile manipulation. However, tracking a 6D object pose distribution on approach can be challenging due to the limited view of the robot camera. In this study, we present a particle filter-based multi-view 6D pose distribution tracking framework that compensates for the limited view of the moving robot camera while it approaches the object by fusing observations from external stationary cameras in the environment. We extend the single-view pose distribution tracking framework (PoseRBPF) to fuse observations from external cameras. We model the object pose posterior as a multi-modal distribution and introduce techniques for fusion, re-sampling, and pose estimation from the tracked distribution to effectively handle noisy and conflicting observations from different cameras. To evaluate our framework, we also contribute a real-world benchmark dataset. Our experiments demonstrate that the proposed framework yields a more accurate quantification of object pose and associated uncertainty than previous research. Finally, we apply our framework for pre-grasp planning on mobile robots, demonstrating its practical utility.