Search Results (78)

With the continuous advancement of marine development, underwater operational tasks are becoming increasingly diverse and complex. Addressing the limitations of traditional methods and intelligent planning—which focus solely on acquiring task skills while separating grasp planning from force planning—this paper proposes a modeling approach integrating impedance control with deep reinforcement learning. Using a five-finger humanoid underwater dexterous hand as the grasping execution platform, this method achieves collaborative decision-making between grasp planning and force control for underwater dexterous hands. First, a modular underwater dexterous grasping scenario is established. Its kinematic model and inverse solution are analyzed, and the grasping problem is modeled as a Markov decision process. Second, based on the dexterous fingertip impedance control model for simulation, a grasping strategy learning method grounded in deep reinforcement learning is constructed to address the complex control challenges posed by the high degrees of freedom of the dexterous manipulator. Finally, the Proximal Policy Optimization (PPO) algorithm is employed for grasping strategy learning. An underwater dexterous grasping parallel training and testing environment is established using the Isaac Lab simulation platform to rapidly validate the learning method. Simulation results demonstrate the proposed method’s excellent dexterous compliant control performance and strong robustness to underwater variable environments: the PPO-based impedance control scheme reduces contact force variance by 56% compared to pure position control. The average maximum contact force is suppressed to 3.26 N, representing a 60.4% reduction compared to pure position control. This study achieves the organic integration of underwater hydrodynamic compensation, adaptive impedance control, and grasping strategy learning, providing a novel and effective solution for compliant grasping control of underwater dexterous manipulators. Full article

This paper presents robotic manipulation methods for rapid high-arc object transfer using dynamic, non-prehensile interactions. Two complementary techniques are introduced, two-fingered scoop-and-flick and one-fingered topple-and-flick, designed for objects with low and high centers of mass, respectively. Both methods enable a robot to retrieve objects resting on a surface and launch them into controlled projectile trajectories without requiring stable grasp formation. To support these maneuvers, we develop physics-based models of object acquisition and release, and combine them with a data-driven framework. While analytical modeling guides the acquisition phase, the highly nonlinear flicking dynamics are captured using learned predictive models that enable accurate selection of control parameters for desired trajectories. The proposed techniques enable dynamic object transfer, reduced grasp planning complexity, and adaptability to environmental constraints. Experiments conducted on a custom robotic platform demonstrate reliable and accurate high-arc object transfer, in which the majority of object displacement is achieved through projectile motion. Full article

This paper presents the design of a high-efficiency apple harvesting robot based on a hybrid pneumatic-electric drive system, capable of operating around the clock. The robotic system comprises a mobile platform with two degrees of freedom (DOF) and a five-DOF PRRRP manipulator for fruit picking. To meet the harvesting requirements, a spoon-shaped end-effector with pneumatic control was developed, enabling precise manipulator control and flexible grasping. The robot’s vision system integrates machine vision and deep neural network approaches. Additionally, an industrial computer and AC servo drivers were employed to control the manipulator and end-effector. An integrated nighttime illumination system allowed for all-weather operation. Initial experiments were conducted in a controlled laboratory. Subsequently, comprehensive identification and harvesting tests were performed in both laboratory and field environments to validate system robustness. Experimental results validated the effectiveness of the proposed system, demonstrating an apple harvesting success rate of 81% and an average harvesting time of 7.81 s per apple. The system achieved a fruit damage rate of less than 5% during field experiments, demonstrating its potential for gentle handling. The primary innovation of this work lies in its hybrid drive architecture and adaptive vision strategy, which together offer a cost-effective and robust solution for all-weather automated harvesting, addressing key limitations of high cost and environmental sensitivity in existing robotic harvesters. Full article

Plant-parasitic nematodes (PPNs) cause big crop losses globally. Safe/reliable methods for their durable management strategies can harness various beneficial relationships among the plant immune system and related microbiomes. Molecular mechanisms basic to these relations reveal wide arrays of significant roles for plant-healthy growth. This review focuses on such relations of microbiomes to prime and immunize plants against PPNs. It also highlights molecular issues facing PPN-resistant varieties with possible solutions such as genetic breeding/engineering, grafting, PPN-antagonistic root exudates, and novel resistant cultivars. These issues call for optimal uses of various widespread groups of microbiomes. Related plant signaling hormones and transcription factors that regulate gene expression and modulate nematode-responsive genes to ease positive/negative adaptation are presented. Exploring PPN-resistance genes, their activation mechanisms, and signaling networks offers a holistic grasp of plant defense related to biotic/abiotic factors. Such factors relevant to systemic acquired resistance (SAR) via plant–microbe interactions to manage PPNs are stressed. The microbiomes can be added as inoculants and/or steering the indigenous rhizosphere ones. Consequently, SAR is mediated by the accumulation of salicylic acid and the subsequent expression of pathogenesis-related genes. To activate SAR, adequate priming and induction of plant defense against PPNs would rely on closely linked factors. They mainly include the engaged microbiome species/strains, plant genotypes, existing fauna/flora, compatibility with other involved biologicals, and methods/rates of the inoculants. To operationalize improved plant resistance and the microbiome’s usage, novel actionable insights for research and field applications are necessary. Synthesis of adequate screening techniques in plant breeding would better use multiple parameters (molecular and classical ones)-based ratings for PPN-host suitability designation. Sound statistical analyses and interpretation approaches can better identify genotypes with high-level, stable resistance to PPNs than the commonly used ones. Linking molecular mechanisms to consistent field relevance can be progressed via dissemination of many advanced techniques. The CRISPR/Cas9 system has been effective in knocking out both the OsHPP04 gene in rice to confer resistance against Meloidogyne graminicola and the GhiMLO3 gene in cotton to minimize the Rotylenchulus reniformis reproduction. Its genetic modifications in crops synthesized “transgene-free” PPN-resistant plants without decreased growth/yield. Characterizing microbiome species/strains needed to prime and immunize plants requires better molecular tools for fine-scale taxonomic resolution than the common ones used. The former can distinguish closely related ones that exhibit divergent phenotypes for key attributes like stability and production of enzymes and secondary metabolites. As PPN-control strategies via tritrophic interactions are more sensitive to the relevant settings than chemical nematicides, it is suggested herein to test these settings on a case-by-case basis to avoid erratic/contradictory results. Moreover, expanding the use of automated systems to expedite detection/count processes of PPN and related microbes with objectivity/accuracy is discussed. When PPNs and their related microbial distribution patterns were modeled, more aspects of their field distributions were discovered in order to optimize their integrated management. Hence, the feasibility of site-specific microbiome application in PPN–hotspot infections can be evaluated. The main technical challenges and controversies in the field are also addressed herein. Their conceptual revision based on harnessing novel techniques/tools is direly needed for future clear trends. This review also engages raising growers’ awareness to leverage such strategies for enhancing plant resistance and advancing the microbiome role. Microbiomes enjoy wide spectrum efficacy, low fitness cost, and inheritance to next generations in durable agriculture. Full article

Robotic grippers play a crucial role in pick-and-place tasks, as their performance directly affects the robot’s operational efficiency, stability, and safety. In industrial applications, such as coal gangue sorting, the target objects have irregular shapes and sharp surfaces, which pose challenges to the gripper’s grasping ability. To solve these problems, an adaptive under-actuated gripper based on rope control is designed. The gripper is simple to control and combines the excellent features of both rigid and flexible grippers. To analyze the characteristics of the gripper, both mathematical analysis and holding force experiments are conducted. The results show that the gripper can generate a greater holding force when grasping larger objects with a constant input air pressure. Furthermore, irregularly shaped testing objects, including coal lumps and ores, are selected to conduct grasping experiments. The gripper achieves a 100% grasping success rate with a load of up to four times the object’s weight suspended beneath it and shows the ability to reliably grasp irregularly shaped objects in high-speed pick-and-place tasks with a payload of four times the object’s weight. Meanwhile, the gripper has a passive anti-collision ability due to the special outer contour of the distal finger when subjected to unexpected, sudden force. Full article

Soft grippers exhibit excellent adaptability in handling objects of various shapes. However, due to the large deformation and high compliance of their constituent materials, the integration of sensing capabilities has long been a major research challenge. Based on the swallowing-type soft gripper proposed in previous work, this study explores the gripper’s capability to perceive object information by leveraging the characteristic that the sealed cavity undergoes volume change due to compression by the object during swallowing, thereby altering the pressure of the internal fluid medium. By establishing the geometric configuration of the sealed cavity composed of elastic membranes, the volume-pressure variation sensing model during the object swallowing process was derived. The performance of this sensing method was tested, and the application of the fluid pressure sensing strategy in closed-loop control was demonstrated, including the classification of objects by shape and sorting by size. This work provides a solution for the object shape-adaptive swallowing-type soft gripper to achieve sensory grasping functionality. Full article

In response to the growing need for efficient, stable, and non-destructive gripping in apple harvesting robots, this study proposes a novel rigid–flexible coupled end-effector. The design integrates an underactuated mechanism with a real-time force feedback control system. First, compression tests on ‘Red Fuji’ apples determined the minimum damage threshold to be 24.33 N. A genetic algorithm (GA) was employed to optimize the geometric parameters of the finger mechanism for uniform force distribution. Subsequently, a rigid–flexible coupled multibody dynamics model was established to simulate the grasping of small (70 mm), medium (80 mm), and large (90 mm) apples. Additionally, a harvesting experimental platform was constructed to verify the performance. Results demonstrated that by limiting the contact force of the distal phalange region silicone (DPRS) to 24 N via active feedback, the peak contact forces on the proximal phalange region silicone (PPRS) and middle phalange region silicone (MPRS) were effectively maintained below the damage threshold across all three sizes. The maximum equivalent stress remained significantly below the fruit’s yield limit, ensuring no mechanical damage occurred, with an average enveloping time of approximately 1.30 s. The experimental data showed strong agreement with the simulation, with a mean absolute percentage error (MAPE) of 5.98% for contact force and 5.40% for enveloping time. These results confirm that the proposed end-effector successfully achieves high adaptability and reliability in non-destructive harvesting, offering a valuable reference for agricultural robotics. Full article

The escalating volume of waste electrical and electronic equipment (WEEE) poses a significant global environmental challenge. The disassembly of printed circuit boards (PCBs), a critical step for resource recovery, remains inefficient due to limitations in the adaptability and dexterity of existing automated systems. This paper proposes an intelligent disassembly system for PCB components that integrates a multimodal large language model (MLLM) with a multi-agent framework. The MLLM serves as the system’s cognitive core, enabling high-level visual-language understanding and task planning by converting images into semantic descriptions and generating disassembly strategies. A state-of-the-art object detection algorithm (YOLOv13) is incorporated to provide fine-grained component localization. This high-level intelligence is seamlessly connected to low-level execution through a multi-agent framework that orchestrates collaborative dual robotic arms. One arm controls a heater for precise solder melting, while the other performs fine “probing-grasping” actions guided by real-time force feedback. Experiments were conducted on 30 decommissioned smart electricity meter PCBs, evaluating the system on recognition rate, capture rate, melting rate, and time consumption for seven component types. Results demonstrate that the system achieved a 100% melting rate across all components and high recognition rates (90–100%), validating its strengths in perception and thermal control. However, the capture rate varied significantly, highlighting the grasping of small, low-profile components as the primary bottleneck. This research presents a significant step towards autonomous, non-destructive e-waste recycling by effectively combining high-level cognitive intelligence with low-level robotic control, while also clearly identifying key areas for future improvement. Full article

Designing a mechanical gripper, achieving the combined capabilities of high loading capacity, flexible environmental adaptability, and dexterous kinematic performance, is highly desired in human–machine interaction and industrial production efficiency improvement, yet this combination of grasping encounters irreconcilable challenges. Although rigid–flexible coupled mechanical grippers exhibit promising advantages compared with conventional rigid mechanical grippers and pure soft grippers, they still get stuck in problems of grasping stability owing to the mechanical mismatch between rigid and flexible materials. Inspired by the hybrid structure of the human finger, we designed an underactuated rigid–flexible coupled mechanical gripper (U-RFCG) to expand the grasping range of existing mechanical grippers. We utilized an embedded flexible microcolumn array to couple the rigid underactuated fingers with a flexible silicone rubber finger segment and integrated a flexible silicone rubber cavity into each rigid–flexible coupling finger segment, thereby addressing issues such as slippage and fracture at the coupling interface of the rigid–flexible structure. This design enables the mechanical gripper to possess the superior characteristics of both rigid and flexible grippers, along with simple execution control. We established mathematical models to analyze the static and kinematic properties of the fingers. Based on these models, we optimized the dimensional parameters of the underactuated links to ensure reasonable contact force distribution and stable motion. Repeated experiments demonstrated that the contact force exerted by each phalanx consistently stabilized at approximately 3.58 N during operation. Lastly, we integrated the U-RFCG into a 3D motion platform. Our mechanical gripper demonstrates significant adaptability and high load capacity for grasping various objects, including irregular cauliflowers, fragile fried instant noodles, and heavy cabbages. It successfully handled objects spanning a weight range of 30–1500 g without causing damage to them. These results confirm that our design balances load capacity and grasping safety through the synergy of rigid and flexible properties, providing a new solution for robotic grasping in complex scenarios. Full article

The chain-of-thought (CoT) approach in large language models (LLMs) has markedly enhanced their performance on complex tasks; however, effectively distilling this capability into LLMs with smaller parameter scales remains a challenge. Studies have found that small LLMs do not always benefit from CoT distillation. Inspired by the concept of teaching students in accordance with their aptitude, we propose an adaptive chain-of-thought distillation (ACoTD) framework. The core idea is to dynamically and adaptively customize distillation data and supervision signals for student models based on their performance on the original problems. Specifically, ACoTD initially evaluates and categorizes the original problems according to the capabilities of the student model. Subsequently, for Easy- and Medium-level problems, a short CoT distillation is employed for a brief lecture to reinforce knowledge and enhance training efficiency, for high-difficulty problems where the student model underperforms, and a detailed long CoT distillation is utilized for in-depth explanation to infuse richer reasoning logic. This differentiated distillation strategy ensures that student models achieve a better grasp of learning. We conducted experiments on multiple benchmark datasets. The results indicate that, compared to the baseline, our method can significantly improve the inference performance of small LLMs. Our method provides a new student-centered paradigm for knowledge distillation, demonstrating that adaptive adjustment of teaching strategies based on student feedback is an effective way to enhance small LLMs’ reasoning ability. Full article

Nowadays, grippers are extensively employed to interact with dynamic and variable objects. Therefore, enhancing the adaptability of grippers is crucial for improving production efficiency and product quality. To address the trade-off between load capacity and interaction safety in rigid and soft grippers, this paper proposes a rigid–flexible coupling gripper with high grasping adaptability based on an underactuated structure. We conduct static analysis on the underactuated mechanism, followed by dimensional optimization using a genetic algorithm. After optimization, the grasping force error at each knuckle is reduced to 2 N, and the total grasping force reaches 38 N. The soft actuators, integrated with a rigid framework, not only increase the contact area during grasping but also mitigate the excessive concentration of contact forces, significantly improving the compliance of the gripper. Additionally, to tackle the issue of weak interfacial bonding strength caused by rigidity mismatch between rigid components and soft materials, this paper proposes a novel method of applying embedded microstructures to enhance the interfacial toughness of rigid–flexible coupling. The elastic deformation of these microstructures ensures strong interfacial connection strength both under tensile and shear stresses. Lastly, a robotic grasping platform is developed to carry out diverse grasping experiments. Experimental results show that the underactuated linkage mechanism and the flexible structure can collaboratively adjust grasping strategies when handling objects of various types, enabling stable manipulation while preventing object damage. This design effectively expands the operational applicability of the gripper. Full article

Modern agricultural intelligent manufacturing faces critical challenges including low automation levels, safety hazards in high-temperature processing, and insufficient production data integration. Digital twin technology and 3D printing offer promising solutions through real-time virtual–physical synchronization and customized equipment manufacturing, respectively. However, existing research exhibits significant limitations: inadequate real-time synchronization mechanisms causing delayed response, poor environmental adaptability in unstructured agricultural settings, and limited human–machine collaboration capabilities. To address these deficiencies, this study develops a digital twin-driven intelligent sorting system for 3D-printed agricultural tools, integrating an Articulated Robot Arm, 16 industrial-grade 3D printers, and the Unity3D 2024.x platform to establish a complete “printing–sorting–warehousing” digitalized production loop. Unlike existing approaches, our system achieves millisecond-level bidirectional physical–virtual synchronization, implements an adaptive grasping algorithm combining force control and thermal sensing for safe high-temperature handling, employs improved RRT-Connect path planning with ellipsoidal constraint sampling, and features AR/VR/MR-based multimodal interaction. Validation testing in real agricultural production environments demonstrates a 98.7% grasping success rate, a 99% reduction in burn accidents, and a 191% sorting efficiency improvement compared to traditional methods, providing breakthrough solutions for sustainable agricultural development and smart farming ecosystem construction. Full article

Efficient production scheduling is a key challenge in industrial operations and continues to attract significant interest within the field of operations research. This paper investigates a range of methodological approaches designed to solve the permutation flow shop scheduling problem (PFSP) with sequence-dependent setup times (SDST). The main objective is to minimize the total weighted flow time (TWFT) while ensuring a no-wait production environment. The proposed solution strategy is based on using algorithms with a mixed integer linear programming (MILP) formulation, heuristics, and their combination. The heuristics utilized in this paper include an advanced greedy randomized adaptive search procedure (GRASP) based on a priority rule and Hybrid-GRASP-NEH (HGRASP), where Nawaz-Enscore-Ham (NEH) takes place to initiate solutions, based on iterative global and local search methods to refine exploration capabilities and improve solution quality. These approaches were validated using a comprehensive set of experiments across diverse instance sizes that proved the efficiency of HGRASP, with the results showing a high-performance level that closely matched that of the exact MILP approach. Statistical analysis via the Friedman test (χ² = 46.75, p = 7.04 × 10⁻¹¹) confirmed significant performance differences among MILP, GRASP, and HGRASP. While MILP guarantees theoretical optimality, its practical effectiveness was limited by imposed computational time constraints, and HGRASP consistently achieved near-optimal solutions with superior computational efficiency, as demonstrated across diverse instance sizes. Full article

Pick-and-place handling of aquatic products (e.g., raw oyster) in packing processing remains manual, despite advances in soft robotic grippers as well as robotic systems that offer a path to automation in food production lines. In this study, we focused on the automation of raw-oyster handling which can be achieved by a robotic system equipped with a soft robotic gripper. However, raw oysters are fragile and prone to large damage during robotic handling, while high-speed handling generates inertial effects. Minimizing the grasping force is thus essential to protect raw oysters, while preserving the grasping stability is required. To address, this study introduces and validates a robotic system equipped with a soft pneumatic gripper for raw-oyster handling task in food production lines. Finite element analysis (FEA) was employed to discuss the effect of gripper actuation pressure on finger deflection and gripper grasping force, revealing a trade-off: increasing actuation pressure improves stability but raises grasping force, whereas reducing actuation pressure causes excessive swing and tossing problems. An optimal actuation pressure of the soft gripper was identified as grasping stability and oyster integrity, minimizing swing while preventing excessive grasping force. Handling performance of the robotic system was experimentally evaluated with raw oysters under different actuation pressures and oyster orientations. Under the optimal actuation pressure confirmed in FEA, the robotic system achieved a handling success rate of 100% (15/15) without obvious misalignment and large damage of raw oysters, which confirmed its adaptability for high-speed, stable handling. This study offers a reference of robotic systems for handling fragile aquatic products and indicates that optimal actuation pressure can protect such products during robotic handling, thereby facilitating the automation of aquatic product processing. Full article

Despite rapid advances in robot learning, the coordination of closed-loop manipulation in cluttered environments remains a challenging and relatively underexplored problem. We present a novel two-level hierarchical architecture for a depth vision-equipped robotic arm that integrates pushing, grasping, and high-level decision making. Central to our approach is a prioritised action–selection mechanism that facilitates efficient early-stage learning via behaviour cloning (BC), while enabling scalable exploration through reinforcement learning (RL). A high-level decision neural network (DNN) selects between grasping and pushing actions, and two low-level action neural networks (ANNs) execute the selected primitive. The DNN is trained with RL, while the ANNs follow a hybrid learning scheme combining BC and RL. Notably, we introduce an automated demonstration generator based on oriented bounding boxes, eliminating the need for manual data collection and enabling precise, reproducible BC training signals. We evaluate our method on a challenging manipulation task involving five closely packed cubic objects. Our system achieves a completion rate (CR) of 100%, an average grasping success (AGS) of 93.1% per completion, and only 7.8 average decisions taken for completion (DTC). Comparative analysis against three baselines—a grasping-only policy, a fixed grasp-then-push sequence, and a cloned demonstration policy—highlights the necessity of dynamic decision making and the efficiency of our hierarchical design. In particular, the baselines yield lower AGS (86.6%) and higher DTC (10.6 and 11.4) scores, underscoring the advantages of content-aware, closed-loop control. These results demonstrate that our architecture supports robust, adaptive manipulation and scalable learning, offering a promising direction for autonomous skill coordination in complex environments. Full article