Robotics and Computer-integrated Manufacturing最新文献

{"title":"Safe multi-channel communication for human–robot collaboration","authors":"Gorkem Anil Al,&nbsp;Uriel Martinez-Hernandez","doi":"10.1016/j.rcim.2025.103109","DOIUrl":"10.1016/j.rcim.2025.103109","url":null,"abstract":"<div><div>This paper presents a safe multi-channel communication and safety system for human–robot collaboration (HRC) in industrial applications enabled by the DiGeTac unit. This unit integrates gesture, distance, and custom-designed tactile sensors, with gesture and distance elements on the top and the tactile element on the bottom. This design provides enhanced multimodal safety and interaction, enabling both close proximity and long-distance perception, making the DiGeTac unit highly suitable for various collaborative scenarios. Unlike other multimodal sensors, DiGeTac offers contactless and touch-based interaction, and post- and pre-collision safety features for a broader range of tasks in HRC environments. The performance of each sensing element within the DiGeTac unit is thoroughly evaluated through a series of validation experiments with a robot arm. The distance sensor’s accuracy is assessed in pre-collision scenarios, ensuring reliable proximity detection for collision avoidance as part of the safety strategy. The tactile sensor is tested in a post-collision scenario, where it functions as a safety mechanism to detect impacts and trigger protective responses. The capability of hand gestures recognition to facilitate intuitive human–robot communication is evaluated using an artificial neural network (ANN). Additionally, the tactile sensor’s contact estimation is analysed with a convolutional neural network (CNN), enhancing the robot’s ability to interact with humans and perform collaborative tasks. Finally, both safety and interaction strategies are tested in HRC scenarios, where the human operator commands the robot to move to specific positions. The results show that the DiGeTac unit is effective and has potential to improve complex collaborative tasks.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103109"},"PeriodicalIF":11.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144922016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TS-RIL: A two-stage robot imitation learning framework with motion trajectory learning and obstacle avoidance in real-world operating scenarios","authors":"Yuming Ning ,&nbsp;Tuanjie Li ,&nbsp;Cong Yao ,&nbsp;Wenqian Du ,&nbsp;Yan Zhang ,&nbsp;Yulin Zhang","doi":"10.1016/j.rcim.2025.103111","DOIUrl":"10.1016/j.rcim.2025.103111","url":null,"abstract":"<div><div>Robot imitation learning is an important research and application direction in the field of robot-based intelligent manufacturing. This technology aims to enhance the autonomous and robustness of robot manipulation in unstructured environments. In this paper, we propose a two-stage robot imitation learning framework with motion trajectory learning and obstacle avoidance, named TS-RIL. Specifically, the proposed TS-RIL consists of a motion trajectory learning algorithm based on the region-constrained dynamic motion primitives (RC-DMPs) and an obstacle avoidance algorithm based on the smooth rapidly-exploring random tree star (S-RRT*), and its basic idea is to divide the robot imitation learning into two stages: motion trajectory learning stage and obstacle avoidance stage. In the motion trajectory learning stage, we design a coupling term based on artificial potential field to extend DMPs to RC-DMPs, and generate the learning trajectories in the constrained regions by real-time updating the learning parameters such as the weights of the forcing term, scaling factor and convergence factor. Then in the obstacle avoidance stage, we propose the S-RRT* algorithm to search for a smooth motion trajectory in the local obstacle region, and recombine it with the trajectory generated by RC-DMPs to generalize a new collision-free motion trajectory. Finally, we further develop the continuous RC-DMPs system architecture, which enables robot trajectory learning, obstacle avoidance and motion execution can be performed sequentially and alternately. To evaluate the overall performance of the proposed TS-RIL, we develop an algorithm verification platform based on the Robot Operating System (ROS) and conduct a series of typical prototype experiments in complex real-world scenarios. The experimental results demonstrate that our TS-RIL can significantly improve the effectiveness and robustness of robot motion trajectory learning and generalization, and outperforms the existing robot motion trajectory learning methods in terms of efficiency, path length and success rate.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103111"},"PeriodicalIF":11.4,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144921298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A framework for integrated design of human–robot collaborative assembly workstations","authors":"Martina Salami ,&nbsp;Pietro Bilancia ,&nbsp;Margherita Peruzzini ,&nbsp;Marcello Pellicciari","doi":"10.1016/j.rcim.2025.103108","DOIUrl":"10.1016/j.rcim.2025.103108","url":null,"abstract":"<div><div>Collaborative robotics is increasingly considered in manufacturing to improve efficiency while reducing operators physical and cognitive workloads. However, the lack of comprehensive methodologies has limited the consistent implementation of human–robot collaborative workstations across industries. Existing approaches are often fragmented, require robotics expertise, and pose challenges for non-experts, leading to suboptimal station designs and inefficient task allocation. This study presents a structured design framework to transition traditional assembly processes into collaborative ones. The framework provides a practical, scalable solution for optimizing collaborative workstations, balancing performance, ergonomics, and industrial applicability. It starts from the analysis of the assembly tasks, followed by classification and allocation between human operators and robots, and concludes with virtual prototyping and performance optimization through simulation using a commercial tool. The adopted methodology integrates task analysis, ergonomic assessment, and workspace design to ensure accessible and efficient implementation. Validated through two industrial case studies involving a gear pump and a worm gearbox, the approach demonstrated significant reductions in cycle time and notable improvements in the ergonomic working conditions. Additionally, physical prototyping and testing conducted within a research collaborative cell further confirmed the achieved results.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103108"},"PeriodicalIF":11.4,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144917926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Probing AR-assisted seamless HRC assembly for industry 5.0: Multi-modal mutual cognition and LLM-driven knowledge reasoning","authors":"Ye Ma,&nbsp;Dunbing Tang,&nbsp;Haihua Zhu,&nbsp;Qixiang Cai,&nbsp;Zequn Zhang,&nbsp;Liping Wang,&nbsp;Changchun Liu","doi":"10.1016/j.rcim.2025.103112","DOIUrl":"10.1016/j.rcim.2025.103112","url":null,"abstract":"<div><div>The advancement of Industry 5.0 fosters human-centric manufacturing, aiming to enhance the well-being and needs of operators. Current research on Human-robot Collaboration (HRC) in the assembly scene is progressively evolving toward intelligent manufacturing with a human-centric focus, particularly in safety and interaction. However, in existing HRC assembly environments, the disjointed relationship between humans and robots presents challenges in handling complex manufacturing tasks. Cobots often struggle to accurately comprehend human actions and assembly contexts, while operators lack real-time insights into the current assembly scene. This mismatch increases the cognitive and physical burden on operators, ultimately reducing assembly efficiency and quality. To address this issue and achieve seamless HRC assembly, this paper proposes an Augmented Reality (AR) assisted HRC assembly method that integrates multimodal mutual cognition with Large Language Model (LLM) driven knowledge reasoning. Firstly, a Transformer-based multimodal fusion perception method for HRC scenarios is proposed to overcome the limitations of current cobots in understanding and responding to human commands and environmental changes in real-time. Based on this, a human-centric mutual cognition and safety framework for HRC is proposed to mitigate interaction risks and promote intelligent coexistence between operators and cobots. Secondly, an LLM-driven assembly knowledge reasoning system is proposed. By constructing an assembly craft information model with semantic associations and dynamic updating capabilities, the system facilitates decision support for HRC tasks and provides optimized assembly plan recommendations. This approach effectively leverages the respective advantages of human operators and cobots in executing assembly tasks. In addition, an AR-based HRC assembly assistance system is designed to enhance visual guidance during the assembly process while integrating the functionalities above. Finally, practical validation is conducted in real-world assembly tasks. Experimental results demonstrate that the proposed method offers efficient, intelligent, and visualized support for HRC assembly tasks, significantly reducing the cognitive load on operators while improving assembly efficiency and product consistency.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103112"},"PeriodicalIF":11.4,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144908403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring tasks and challenges in human-robot collaborative systems: A review","authors":"Seemal Asif ,&nbsp;Tiziana C. Callari ,&nbsp;Fahad Khan ,&nbsp;Iveta Eimontaite ,&nbsp;Ella-Mae Hubbard ,&nbsp;Masoud S. Bahraini ,&nbsp;Phil Webb ,&nbsp;Niels Lohse","doi":"10.1016/j.rcim.2025.103102","DOIUrl":"10.1016/j.rcim.2025.103102","url":null,"abstract":"<div><div>This paper presents an in-depth exploration of Human-Robot Collaborative Systems (HRCSs) within industrial environments, a dynamic field that has witnessed significant advancements due to technological innovation and the increasing integration of Artificial Intelligence (AI). As industries evolve towards more collaborative and adaptive manufacturing systems, the dynamic interaction between humans and robots becomes pivotal. This study reviews the current state of HRCSs, focusing on the challenges of task allocation and skill alignment, safety, trust, and the psychological wellbeing of human workers. We review control strategies and architectural frameworks that underpin effective human-robot interactions (HRI), emphasising the critical role of AI in enhancing decision-making processes and the adaptability of collaborative efforts. Our review sheds light on the complexities involved in designing HRCSs that are not only efficient but also cognisant of the human experience, advocating for a balanced approach that leverages the strengths of both human and robotic counterparts. We argue that research in and implications of HRCSs should extend beyond technical considerations, touching on ethical, social, and organisational dimensions, thereby contributing to the broader discourse on the future of work in the era of Industry 4.0 and future Industry 5.0.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103102"},"PeriodicalIF":11.4,"publicationDate":"2025-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144889434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A cloud manufacturing service composition optimization method for fuzzy demands based on improved NSGA-III algorithm","authors":"Xianhui Liu ,&nbsp;Run Yang ,&nbsp;Xiaobin Li ,&nbsp;Xi Vincent Wang","doi":"10.1016/j.rcim.2025.103106","DOIUrl":"10.1016/j.rcim.2025.103106","url":null,"abstract":"<div><div>The Industrial Internet integrates industrial systems with advanced Internet technologies to establish an intelligent implementation platform and diversified service ecosystem for cloud manufacturing. However, the extensive user base introduces substantial uncertainties in temporal, financial, and operational requirements of cloud manufacturing tasks. While existing studies propose various solutions, their reliance on subjective criteria for demand variation analysis leads to inadequate handling of fuzzy demands. A novel fuzzy demand-oriented optimization method is proposed for cloud manufacturing service composition, employing fuzzy sets and membership functions to establish an objective quantification framework for modeling uncertain demands. The approach formulates a multi-objective optimization model incorporating four key metrics: service cost, service time, service quality, and resource utilization rate, with fuzzy satisfaction functions constructing constraints containing random variables to ensure robust realization of fuzzy demands. An enhanced NSGA-III algorithm is developed featuring opposition-based learning mechanisms and GKM-based reference point generation to enhance population diversity and convergence efficiency. Validation through benchmark functions and practical cloud manufacturing scenarios confirms the method’s effectiveness in addressing fuzzy demand challenges, with the co-evolution algorithm demonstrating superior convergence and diversity performance.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103106"},"PeriodicalIF":11.4,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144866118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robot digital twin systems in manufacturing: Technologies, applications, trends and challenges","authors":"Qiang Qin ,&nbsp;Zhihao Liu ,&nbsp;Ruirui Zhong ,&nbsp;Xi Vincent Wang ,&nbsp;Lihui Wang ,&nbsp;Magnus Wiktorsson ,&nbsp;Wei Wang","doi":"10.1016/j.rcim.2025.103103","DOIUrl":"10.1016/j.rcim.2025.103103","url":null,"abstract":"<div><div>The manufacturing industry is undergoing a profound transformation toward smart, digital, and flexible production systems under the Industry 4.0 framework. Within this paradigm, Digital Twin (DT) serves as a key enabler, bridging physical and digital domains to simulate, analyse, and optimise manufacturing operations. Concurrently, robotic systems, enhanced by smart sensor perception, Industrial Internet of Things connectivity, and adaptive control mechanisms, are increasingly deployed to handle complex and dynamic tasks. However, the evolving demands of the modern manufacturing industry require a high degree of flexibility and responsiveness, necessitating more intelligent solutions. The Robot Digital Twin (RDT) has emerged as a transformative approach, facilitating dynamic adaptation and continuous operational improvement. This review offers a comprehensive examination of the literature on RDT in manufacturing from both technology and application perspectives, aiming to provide insight for researchers and practitioners in Industry 4.0. The paper introduces a four-layer RDT system architecture and summarises how Industry 4.0 technologies, e.g., the Industrial Internet of Things, Cloud/Edge Computing, 5 G, Virtual Reality, Modelling and Simulation, and Artificial Intelligence, converge and influence the RDT system based on this architecture. Furthermore, the review covers domain-specific and system-level applications, such as assembly, machining, grasping, material handling, human-robot interaction, predictive maintenance, and additive manufacturing systems, with an analysis of their development status. Finally, the trends, practical challenges, and future research directions for RDT systems in manufacturing are summarised at different levels.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103103"},"PeriodicalIF":11.4,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cognitive digital twins for capability matching toward reconfigurable manufacturing: Leveraging asset administration shells and large language models","authors":"Dachuan Shi ,&nbsp;Olga Meyer ,&nbsp;Zhi Fan ,&nbsp;Hao Wang ,&nbsp;Thomas Bauernhansl","doi":"10.1016/j.rcim.2025.103105","DOIUrl":"10.1016/j.rcim.2025.103105","url":null,"abstract":"<div><div>Reconfigurable manufacturing (RM) has emerged to support mass customization, which leads to frequent changes in production processes. RM necessitates the rapid reallocation of production resources to accommodate these evolving demands. To address this challenge, we propose a cognitive digital twin (CDT) system that integrates Asset Administration Shells (AAS) and large language models (LLMs) for adaptively matching between production processes and resource capabilities. Our approach centers on the structured representation of knowledge related to products, processes, and resources (PPR) using the AAS and leveraging this foundation for capability matching through the LLM. First, a methodology for developing interoperable AAS submodels (SM) is represented. Based on this, the SM templates of PPR are developed, serving as the knowledge base of the CDT. Next, we propose a capability matching mechanism using the LLM with chain-of-thought prompting. Finally, we design and implement an IT architecture that integrates an LLM-based retrieval-augmented generation system for executing capability matching alongside an AAS server for hosting AAS instances with dynamic values. The proposed CDT system enables the dynamic allocation of production resources to process steps, and is demonstrated and evaluated in a machining center use case. It effectively supports planning customized machining tasks through AAS-based knowledge representation and LLM-powered capability matching.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103105"},"PeriodicalIF":11.4,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robotic pre-drilled Z-pinning approach for enhancing the performance of composite fan blades in aircraft engines","authors":"Zhanxi Wang ,&nbsp;Pengfei Liu ,&nbsp;Xiaoyu Zhang ,&nbsp;Weiwei Liu ,&nbsp;Wenjie Chen ,&nbsp;Xinliang Wang ,&nbsp;Mingxin Yin ,&nbsp;Zhengle Liao ,&nbsp;Xiansheng Qin ,&nbsp;Chen Zheng","doi":"10.1016/j.rcim.2025.103104","DOIUrl":"10.1016/j.rcim.2025.103104","url":null,"abstract":"<div><div>Composite fan blades serve as the primary barrier of protection in aircraft engines during the air intake process, and their mechanical properties play a significant role in aviation safety. The interlaminar performance of composite fan blades is relatively weak owing to the layered manufacturing process, rendering them susceptible to delamination under the application of external impact loads. Z-pin insertion is a technique that involves embedding an array of slender and rigid pins through the laminate to reinforce the composite materials by introducing an additional load-bearing medium through their thickness. However, the application of the Z-pin insertion technique presents significant challenges owing to their complex geometries and stringent performance requirements of the composite fan blades in aircraft engines. The intricate shapes of fan blades require the manual execution of the Z-pin insertion process, leading to low production efficiency and consistency. Moreover, the Z-pins that are not embedded along the normal direction of the curved surface can induce defects, such as matrix cracking or fiber misalignment, particularly on curved or irregular fan blade surfaces. Combined with the high stresses and dynamic loads experienced by fan blades during operation, the aforementioned factors complicate the selection of the optimal Z-pin insertion parameters in advanced aerospace applications. This paper presents a robotic pre-drilled Z-pinning (PDZP) approach for improving the performance of composite fan blades in aircraft engines. A series of experiments, including double cantilever beam and end-notched flexure tests, were conducted to obtain the optimal insertion parameters for the robotic PDZP process. Furthermore, a robotic PDZP posture model was developed to embed the Z-pins along the normal direction and to avoid matrix cracking or fiber misalignment in the curved or irregular regions of the fan blade. Additionally, a compensation method was proposed for the normal deviation during the PDZP process. Finally, a prototype of the robotic PDZP system was developed and subjected to impact tests, which demonstrated that the impact resistance of the composite fan blade satisfied the expected requirements.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103104"},"PeriodicalIF":11.4,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780155","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep reinforcement learning driven by heuristics with Petri nets for enhancing real-time scheduling in robotic job shops","authors":"Sijia Yi ,&nbsp;Jiliang Luo","doi":"10.1016/j.rcim.2025.103097","DOIUrl":"10.1016/j.rcim.2025.103097","url":null,"abstract":"<div><div>In robotic job shops (RJS), a significant challenge lies in optimizing task allocation and robot routing simultaneously, especially since these tasks must be accomplished in real-time to efficiently manage unexpected situations, such as the urgent need for AGV recharging or sudden order additions. Deep reinforcement learning (DRL) shows promise for these complex scheduling tasks due to its ability to address problems characterized by substantial computational complexity. However, the rapid expansion of RJS state space and the difficulty of avoiding cyclic loops for AGVs pose significant challenges for DRL in realistic settings. To address these, we present a novel approach combining an artificial-potential-field (APF) with a deep Q-network (DQN) in a Petri net framework. The APF is designed for Petri nets to guide token movement toward goal place nodes. Throughout the learning process, the APF-guided mixed policy employs a cosine-annealing probability for APF policy and a piecewise linear probability for random policy. Initially, action selections predominantly rely on APF policy to efficiently gather high-reward experience. As training progresses, they shifts to more rely on the learned neural-network policy, with random exploration supplementing diversity, ensuring a robust transition from reward-driven exploration to precise decision-making. The APF-DQN method is tested in real-world RJS scenarios, showing superior exploration success and training efficiency over baseline DQN. It significantly outperforms both conventional dispatching rules and baseline DQN, reducing average makespan by over 55% compared to dispatching rules and by 14.9% relative to baseline DQN. This method significantly enhances traditional DQN by improving exploration success, learning efficiency, policy convergence, and adaptability to dynamic environments.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"97 ","pages":"Article 103097"},"PeriodicalIF":11.4,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}