Search Results (181)

Modern manufacturing faces increasing demands for flexibility, customization, and productivity under dynamic conditions. Multi-robot systems offer a promising solution by enabling cooperative execution of complex tasks, such as assembly and cooperative manipulation. In this context, Multi-Agent Reinforcement Learning (MARL) has emerged as a promising paradigm to enhance coordination and adaptability in industrial settings. MARL enables multiple agents to learn and interact in shared environments to achieve common goals within complex and dynamic industrial processes. In this paper, a deep analysis of MARL applied to industrial multi-robot systems based on a systematic review is presented, with particular focus on cooperative manipulation tasks. Following PRISMA guidelines, we analyze a total of 30 articles published between 2016 and 2026, selected independently by two of the authors from an initial pool of 102 records retrieved from Scopus and Web of Science. These articles were used to address five key questions regarding MARL algorithms, control architectures, industrial applications and validation practices. These research questions seek to examine gaps and trends at the research level which are important for the development of multi-agent control technologies. This review shows a clear prevalence of model-free algorithms under Centralized Training with Decentralized Execution (CTDE) architectures, with validation mainly performed in simulation. Despite promising results and high potential for impact, critical gaps remain in scalability, reproducibility, and sim-to-real transfer, limiting real deployment in manufacturing environments. To address these challenges and fill current gaps, we outline actionable research directions, such as hybrid MARL approaches, standardized industrial benchmarks, digital twin pipelines, and safety-aware deployment strategies, to accelerate MARL adoption in industrial environments. Full article

Three-dimensional point cloud understanding plays an important role in autonomous perception, robotic navigation, and LiDAR-based remote sensing. However, the irregular and unordered nature of point clouds makes it challenging to model long-range contextual dependencies while preserving geometric awareness, particularly under noise, occlusion, and non-uniform sampling. To address these challenges, we propose a Geometry-Aware Contextual Memory Network (GACM-Net) for 3D point cloud analysis. Specifically, an Adaptive Geometric Prior Encoding (AdGPE) module is introduced to dynamically coordinate multiple geometric priors, including absolute coordinates, center-relative coordinates, and distance-based cues, thereby enhancing geometry-aware contextual interaction during long-range propagation. Furthermore, a Structure-Aware Point Memory Unit (SAPM) is designed to achieve stable contextual memory learning through normalized gate interaction, peephole memory regulation, gated candidate filtering, and residual feature propagation. Based on SAPM, a Bidirectional Structure-Aware Point Memory module (BiSAPM) further captures complementary geometric dependencies from opposite propagation directions, improving contextual completeness and structural consistency for irregular point cloud representations. Extensive experiments on ModelNet40, ScanObjectNN, and ShapeNetPart demonstrate that GACM-Net achieves competitive classification and part segmentation performance with a compact model size and favorable computational efficiency. The results further show that the proposed framework provides a good balance among accuracy, efficiency, and robustness for 3D point cloud understanding. Full article

Multi-agent deep reinforcement learning (MARL) extends deep reinforcement learning (DRL) to environments involving multiple interacting agents and has enabled applications in domains such as autonomous vehicles, robotics, unmanned aerial vehicles (UAVs), and multi-player games. Compared with single-agent learning, MARL introduces additional challenges, including non-stationarity, partial observability, multi-agent credit assignment, and scalability. This paper presents a narrative survey of recent developments in MARL and discusses major approaches proposed to address these challenges. In particular, we examine research directions centred on centralised training with decentralised execution (CTDE), value decomposition, learned communication, graph-based methods, and model-based learning. We further discuss commonly used benchmark environments and evaluation practices, highlighting considerations related to reproducibility, robustness, and generalisation. Finally, we outline open research challenges and future directions concerning theoretical understanding, sample efficiency, scalable coordination, and deployment in real-world settings. Rather than providing an exhaustive systematic review, this survey aims to offer an organised and up-to-date synthesis of recent progress in MARL. Full article

This survey presents a large-scale, reproducible, and parameter-mapped benchmarking analysis of path planning algorithms for multiple mobile robot systems (MMRS) by systematically examining 247 rigorously filtered papers from high-impact journals. Unlike prior reviews that primarily provide conceptual taxonomies, this survey introduces execution-oriented multi-parameter mapping enabling direct comparison of classical planners (A*, D*, Cell Decomposition, APF, RM, RRT, and ORCA), nature-inspired metaheuristics (PSO, GA, ACO, GWO, FA, ABC, BFO, CS, BA, SFLA, eagle-inspired optimizers), and learning-driven AI frameworks (Fuzzy Logic, Artificial Neural Networks, and Deep Reinforcement Learning). Each paper is evaluated across 15 practical planning dimensions, including environment type (static 95% vs. dynamic 51%), multi-robot validation (52%), dynamic goal handling (13%), energy awareness (14%), timepath optimization bias (82% focus), inter-robot coordination (less than 47%), and software validation platforms (MATLAB 42% and ROS 9%), revealing that simulation-only validation dominates (98%) while experimental testing remains limited (33%). Multivariate validation through Multiple Correspondence Analysis further confirms that coordination maturity, energy awareness, and multi-robot applicability are the primary structural differentiators of deployment readiness across algorithm families. The findings emphasize the need for hybrid, energy-aware, and coordination-driven MRPP frameworks supported by experimental benchmarking and reproducible deployment pipelines to advance real-world MMRS autonomy. Full article

Decentralized multi-robot navigation in grid-based industrial environments must reach goals and avoid collisions without centralized control or direct communication. We study a hybrid framework pairing an offline Rapidly-Exploring Random Tree (RRT) expert with a Behavior Cloning (BC) local policy and route reuse, evaluated in a reproducible simulator; collisions are predicted blocked-move events, not physical contacts. Our central result is a quantitative analysis of environment-specific route-library adaptation: a route library generated for one map and deployed on another leaves robots blocked by unfamiliar obstacles, whereas regenerating it on the deployment map cuts collisions by 38–85% and task failures by 43–70% across two- to ten-robot fleets, and replicates on a third corridor layout (30 seeds; Mann–Whitney $p<{10}^{-5}$; permutation $p_{\mathrm{perm}}<0.001$). Simpler remedies, filtering or repairing invalid routes, recover most of this gain, and a non-learning scripted connector matches the trained policy: the effect lives in the route library itself. A centralized prioritized-planning baseline bounds all communication-free variants from above; capping the online RRT baseline’s planning budget preserves its per-task collision quality but sharply raises task failures. A collision-history-sharing add-on coordinating through a shared map rather than messaging gives a limited, layout-dependent benefit not surviving multiple-comparison correction. Full article

This study presents the innovative design and development of biomimetic origami-based soft robotic grippers capable of two-stage grasping. Inspired by the biological structure of the sea urchin mouth, which combines external rigid teeth with an internal soft membrane, the proposed grippers employ origami architectures to achieve coordinated two-stage grasping. Novel waterbomb and Miura-ori origami architectures were introduced, enabling the formation of external and internal teeth. The developed grippers integrate an elastomeric membrane with an internal origami structure that enables contraction-driven folding under negative-pressure actuation. Multiple gripper configurations with varying dimensions are fabricated using paper and polymer-laminated paper skeletons. An energy-based modeling framework is introduced to describe the pressure–force relationship while accounting for the effects of structural deformation. Experimental evaluations conducted at different negative-pressure values quantified grasping performance and holding force. Imprint-based analysis confirmed the two-stage grasping mechanism, while grasping capability investigations demonstrated compliant interaction with delicate objects. Holding forces were measured using cylindrical metal and spherical wooden test objects of varying sizes and orientations. The waterbomb-based gripper achieved the most consistent performance, particularly for cylindrical objects, reaching a maximum holding force of 70 N, whereas the Miura-ori provided improved adaptability and higher holding forces for spherical objects, reaching 74.8 N, and maximum force-to-weight ratios of 327.2 and 346.6 were achieved for the waterbomb- and Miura-ori-based grippers, respectively. Full article

This paper investigates the conflicting multiple constraints and safety challenges in humanoid robot teleoperation for nonprehensile transportation tasks. The robot’s complex workspace and high degrees of freedom frequently conflict with highly dynamic task requirements, imposing stringent demands on coordinated motion. To address these issues, this paper proposes a Multiple-Constraint Safety-Critical Control Framework (MC-SCCF) featuring a hierarchical three-layer architecture. The top layer guarantees intrinsic safety against workspace boundaries using a continuously differentiable reachability surrogate model and an improved control barrier function (CBF)-based safe velocity filter for smooth deceleration. The middle layer maps user commands into pose-coupled reference trajectories to ensure task-level object safety, satisfying strict non-slip and non-toppling constraints. The bottom layer utilizes a quadratic programming (QP)-based inverse kinematics solver to achieve self-collision avoidance, coordinated motion, and optimal configuration while strictly enforcing joint and manipulability limits. Simulations and hardware experiments demonstrate that the MC-SCCF achieves real-time, high-precision reachability evaluation and successfully coordinates task dynamics with physical constraints, enhancing operational safety and the human–robot interaction experience. Full article

Path planning is critical for multi-robot systems (MRS), directly affecting the operation efficiency, execution time, and operational cost. Despite extensive research and successful applications of multiple algorithms, achieving globally optimal solutions in cluttered or dynamic environments remains a significant challenge. Issues such as scalability with an increasing number of robots, computational efficiency, system robustness, and coordination complexity continue to drive the development of more reliable approaches. This study reviews modelling approaches, optimisation criteria, and solution algorithms based on the roadmap planning methods that are widely used for multi-robot path planning (MRPP). It focuses on three graph-based algorithms: MRPP algorithm, central algorithm (CA), and the optimisation central algorithm (OCA). These algorithms utilise visibility graphs (VG) for environment representation and Dijkstra’s algorithm for shortest path computation, while incorporating algebraic connectivity to improve coordination, safety, and scalability. In addition, the technological context and implementation platforms, including simulation environments, cloud robotics, and AI-based frameworks, are conceptually examined. The potential applications of these methods in assistive robotics are highlighted, particularly in supporting a safe and reliable navigation in healthcare and human-centred environments. The article synthesises theoretical and practical insights, identifies current limitations and challenges, and outlines future research directions for efficient, scalable, and robust MRPP. Full article

Contemporary neuroscience has generated extensive empirical insights into perception, memory, prediction, valuation, and consciousness. However, it still lacks an explicit operational architecture capable of explaining how these processes emerge from a unified computational mechanism. This work introduces DIME (Detect–Integrate–Mark–Execute), a unified operational architecture in which perception, memory, valuation, and conscious access are treated as components of a single recurrent computational cycle. The framework is organized around four core elements: engrams, defined as distributed recurrent neural structures that support multiple activation trajectories rather than static memory traces; execution threads, representing temporally extended, causally coherent trajectories of neural activity; marker systems, corresponding to neuromodulatory and limbic mechanisms that regulate value, selection, plasticity, and trajectory competition; and hyperengrams, large-scale integrative states associated with global coordination and conscious access. Within this formulation, DIME provides a mapping between local neural assemblies, temporal sequence dynamics, value-based modulation, and large-scale network integration. Rather than treating perception, memory, and decision-making as partially independent processes, the framework interprets them as different expressions of a single operational loop acting across multiple spatial and temporal scales. The proposed architecture is consistent with empirical findings on hippocampal indexing, recurrent cortical processing, neuromodulatory control, and large-scale network dynamics, while remaining sufficiently general to support applications in artificial intelligence and robotics. Unlike frameworks centered on prediction, memory storage, or global broadcasting, DIME proposes that cognition arises from the recurrent interaction between executable representational structures, trajectory-based processing, value-guided selection, and dynamic large-scale integration. The framework generates explicit and falsifiable predictions regarding context-dependent neural trajectories, marker-mediated state transitions, and large-scale network reconfiguration. In this sense, DIME is not intended as a metaphorical synthesis, but as a testable architectural hypothesis for neuroscience and biologically inspired cognitive systems. Beyond theoretical neuroscience, the framework is also positioned as a transferable design-level reference model for adaptive AI systems, autonomous robotics, and cognitively informed engineering architectures operating in dynamic environments. Full article

This paper describes the development of an automated solid waste sorting system that integrates advanced computer vision pipelines with a robotic manipulator for real-time classification and actuation. The system consists of a Deep Neural Network (DNN) and a YOLOv8-based perception module. Thedeveloped model is capable of accurately detecting and classifying objects with confidence scores exceeding 0.71, and the overall system attained a sorting accuracy of approximately 81.8% across multiple test batches. From an integration perspective, the coordination among the Intel RealSense camera, Raspberry Pi 5, Arduino Uno, ultrasonic sensors, relay-switching circuit, and SCORBOT-ER 4U robotic arm demonstrated reliable communication and execution, enabling accurate pick-and-place operations. Overall, the results confirm that the proposed system provides a functional and scalable proof of concept for automated waste segregation in controlled environments. The study highlights that while current performance is sufficient for low-speed applications, further improvements in dataset diversity, perception robustness, mechanical gripping, and feedback control are necessary to achieve higher accuracy, reliability, and industrial applicability. Full article

Despite recent advances in healthcare robotics, most existing systems remain limited to single-purpose functions and lack the flexibility to collaborate dynamically with clinicians and facility systems. To address these limitations, this study presents an LLM-orchestrated framework for a multifunctional Robotic Health Attendant (RHA) that enables robot actions and environment interactions to be coordinated in healthcare environments. Within this framework, the RHA functions as a multifunctional nursing assistant capable of performing physical, communicative, and informational tasks through natural-language interaction. Tasks are expressed in natural language and decomposed into coordinated behaviors across three functional branches: physical, for navigation, object manipulation, and delivering medication; communicational, for dialog with patients and clinicians; and informational, for retrieving and summarizing clinical knowledge, such as patient education on complex heart transplant procedures. The framework integrates multiple Large Language Models (LLMs) and sensing nodes to combine facility data, patient information, and clinician commands, enabling robots and building systems to act in a context-aware manner through coordinated task execution across robotic and environmental components. Implemented in a simulated environment, the framework demonstrates the feasibility of executing representative tasks through LLM-based orchestration, serving as a proof-of-concept toward integrated robotic assistance in healthcare settings. Full article

This article presents a modeling and cooperative kinematic control framework for heterogeneous mobile manipulators operating in a shared task space. The proposed approach integrates systems with different kinematic structures into a unified pose–shape representation, derived from individual and cooperative Jacobian models, enabling coordinated motion under a common formulation and extendable to multiple robots through a hierarchical architecture. The control strategy exploits system redundancy via null-space projection to incorporate secondary objectives without affecting the primary task. In particular, collision-free obstacle avoidance of the mobile bases and safe joint configuration of the robotic arms are achieved simultaneously while preserving formation tracking. The stability of the cooperative system is established using Lyapunov theory, ensuring asymptotic convergence of tracking errors. The proposed method is validated through numerical simulations in MATLAB under two representative scenarios, demonstrating its capability to handle heterogeneous configurations, maintain coordination, and execute safe and scalable cooperative behaviors. Full article

This study presents the design and evaluation of a markerless 3D scanning system based on the Helios2 time-of-flight 3D camera, with automatic alignment of multiple scans acquired from different viewpoints. The proposed system integrates data acquisition, raw coordinate conversion, point cloud preprocessing, including voxel downsampling, statistical outlier removal, and surface normal estimation using Open3D and a coarse-to-fine automatic registration strategy combining FPFH descriptor matching with RANSAC initialization and ICP refinement, followed by voxel-based fusion of aligned views. The system was experimentally evaluated on three cardboard filament-box objects representing a realistic robotic handling scenario. The reconstructed models were compared against nominal reference data obtained from a commercial Revopoint Miraco markerless scanner using surface deviation analysis. The evaluation yielded an overall mean surface deviation of 0.04 mm and a standard deviation of 1.34 mm, demonstrating that the proposed innovative Helios2 workflow can produce markerless multi-view reconstructions quantitatively comparable to those of an existing commercial scanning device. The results confirm the practical feasibility of automatic markerless alignment for ToF 3D scanning and indicate the suitability of the developed system for robotic perception applications where moderate geometric accuracy and reduced scene preparation are prioritized over ultra-high metrological precision. Full article

This paper presents a novel distributed coordination framework for multi-agent robotic swarms tailored for smart transportation applications. The proposed approach addresses the critical pre-transportation phase where a fleet of mobile robots, eventually with different sizes, must converge to fixed positions around an object to ensure effective caging within a user-defined prescribed time. By leveraging a time-varying diffeomorphic mapping based on an affine transformation, the strategy embeds prescribed-time guarantees within a swarm-inspired framework that maps agents between virtual and real reference frames. This methodology ensures the simultaneous achievement of precise target convergence, finite-time stability regardless of initial conditions, and inherent collision avoidance by explicitly considering the physical footprint of each robotic unit. The control protocol is first derived for scalar systems and subsequently extended to multidimensional robotic fleets using additional diffeomorphism-based techniques, which allow for the management of multiple non-interacting swarms to reduce network communication overhead. Full article

This study aims to enhance the precision of humanoid robots in imitating complex human “walking–grasping” coordinated movements. Addressing limitations in sample efficiency and reward function design in Generative Adversarial Imitation Learning (GAIL), we propose the Similarity Reward-Augmented Generative Adversarial Imitation Learning (SRA-GAIL) framework. The method integrates plantar thin-film resistive pressure sensors to measure the real-time pressure distribution at four key points on both feet, combined with roll/pitch angle data acquired from JY901S inertial measurement units (IMUs). A Lagrangian constraint optimization strategy is employed to achieve gait stability control based on the zero moment point (ZMP). Simultaneously, a visual similarity evaluation module is established using human demonstration trajectories captured by a Logitech C920E camera, augmented by grip force feedback from flexible thin-film pressure sensors on the hands. This enables the design of a multimodal sensor-fused similarity reward function. By incorporating Lagrangian constraint optimization and a maximum entropy reinforcement learning framework, Similarity Reward-Augmented Generative Adversarial Imitation Learning synchronously optimizes gait stability control—guided by zero moment point (ZMP) and roll/pitch data—and vision-based trajectory similarity evaluation. These components address motion stability constraints and trajectory similarity metrics, respectively, generating biomechanically plausible gait strategies. A spatiotemporal attention mechanism parses human motion trajectory features to drive the end-effector for high-precision trajectory tracking. To validate the proposed method, an imitation learning experimental system was constructed on a physical XIAOLI humanoid robot platform, integrating inertial measurement units (IMUs), plantar pressure sensors, and a vision system. Quantitative evaluations were conducted across multiple dimensions, including robot platform analysis, walking stability, object grasping success rates, and end-effector trajectory similarity. The results demonstrate that, compared to Generative Adversarial Imitation Learning (GAIL) and behavioral cloning, Similarity Reward-Augmented Generative Adversarial Imitation Learning achieves a stable object grasping success rate of 93.7% in complex environments, with a 23.8% improvement in sample efficiency. The method maintains a 96.5% compliance rate for zero moment point (ZMP) trajectories within the support polygon, significantly outperforming baseline approaches. This effectively addresses the bottleneck in robot policies adapting to dynamic changes in real-world environments. Full article