Robotics and Autonomous Systems最新文献

SA-STGCN: Structural-Adaptive Spatio-Temporal Graph Convolution with Spatio-Temporal Attunement for skeleton-based gesture recognition 基于骨架的手势识别的时空调谐结构自适应时空图卷积

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-28 DOI: 10.1016/j.robot.2026.105371

Junhui Li, Mohammed A.A. Al-qaness

To enable intuitive and reliable human–robot collaboration, robots must understand human actions at a structural level, making skeleton-based gesture recognition (SGR) a crucial source of precise and robust intention cues. Graph convolutional networks (GCNs) have become a key technology in SGR due to their efficient processing of non-Euclidean data. However, existing methods typically choose between a fixed anatomical prior graph and a fully adaptive dynamic graph, which limits the model’s ability to capture structural invariance and dynamic variability in hand motion simultaneously. To address this challenge, we propose the Structural-Adaptive Spatio-Temporal GCN (SA-STGCN), which relies on an innovative spatiotemporal feature extraction mechanism designed to fuse structural priors with motion-adaptive topology synergistically. Spatially, our designed Spatio-Temporal Attunement (STA) Block integrates two key components in parallel: Relational Semantics Graph Convolution (RS-GC), which constructs a rich structured representation by modeling multiple priors such as physical connectivity, symmetry relationships, and functional groupings, while aggregating features at both the joint and component levels. Meanwhile, Motion Signature Graph Convolution (MS-GC) learns a dynamic, instance-specific topological graph from the data to capture instantaneous motion patterns. Temporally, the Temporal Multi-Scale Aggregation (TMA) Module effectively captures fine-grained motion at varying rates through multi-way dilated convolutions, and the Temporal Saliency Modulator (TSM) further enhances the feature weights of keyframes. These improvements significantly enhance the accuracy and efficiency of GR. The experimental results demonstrate that our model achieves an accuracy of 97.62% on the 14-class task and 95.36% on the 28-class task of the SHREC’17 Track dataset, as well as 93.22% on the FPHA dataset.

为了实现直观和可靠的人机协作，机器人必须在结构层面上理解人类的行为，使基于骨骼的手势识别（SGR）成为精确和强大的意图线索的重要来源。图卷积网络（GCNs）因其对非欧几里得数据的高效处理而成为SGR的关键技术。然而，现有方法通常选择固定的解剖先验图和完全自适应的动态图，这限制了模型同时捕捉手部运动的结构不变性和动态变异性的能力。为了解决这一挑战，我们提出了结构自适应时空GCN (SA-STGCN)，它依赖于一种创新的时空特征提取机制，旨在将结构先验与运动自适应拓扑协同融合。在空间上，我们设计的时空调谐（STA）块并行集成了两个关键组件：关系语义图卷积（RS-GC），它通过建模多个先验（如物理连接、对称关系和功能分组）构建了丰富的结构化表示，同时聚合了关节和组件级别的特征。同时，运动签名图卷积（MS-GC）从数据中学习动态的、特定实例的拓扑图，以捕获瞬时运动模式。在时间上，时间多尺度聚合（TMA）模块通过多路扩张卷积有效捕获不同速率的细粒度运动，时间显著性调制器（TSM）进一步增强关键帧的特征权重。实验结果表明，该模型在SHREC ' 17 Track数据集的14类任务和28类任务上的准确率分别为97.62%和95.36%，在FPHA数据集上的准确率分别为93.22%。

{"title":"SA-STGCN: Structural-Adaptive Spatio-Temporal Graph Convolution with Spatio-Temporal Attunement for skeleton-based gesture recognition","authors":"Junhui Li, Mohammed A.A. Al-qaness","doi":"10.1016/j.robot.2026.105371","DOIUrl":"10.1016/j.robot.2026.105371","url":null,"abstract":"<div><div>To enable intuitive and reliable human–robot collaboration, robots must understand human actions at a structural level, making skeleton-based gesture recognition (SGR) a crucial source of precise and robust intention cues. Graph convolutional networks (GCNs) have become a key technology in SGR due to their efficient processing of non-Euclidean data. However, existing methods typically choose between a fixed anatomical prior graph and a fully adaptive dynamic graph, which limits the model’s ability to capture structural invariance and dynamic variability in hand motion simultaneously. To address this challenge, we propose the Structural-Adaptive Spatio-Temporal GCN (SA-STGCN), which relies on an innovative spatiotemporal feature extraction mechanism designed to fuse structural priors with motion-adaptive topology synergistically. Spatially, our designed Spatio-Temporal Attunement (STA) Block integrates two key components in parallel: Relational Semantics Graph Convolution (RS-GC), which constructs a rich structured representation by modeling multiple priors such as physical connectivity, symmetry relationships, and functional groupings, while aggregating features at both the joint and component levels. Meanwhile, Motion Signature Graph Convolution (MS-GC) learns a dynamic, instance-specific topological graph from the data to capture instantaneous motion patterns. Temporally, the Temporal Multi-Scale Aggregation (TMA) Module effectively captures fine-grained motion at varying rates through multi-way dilated convolutions, and the Temporal Saliency Modulator (TSM) further enhances the feature weights of keyframes. These improvements significantly enhance the accuracy and efficiency of GR. The experimental results demonstrate that our model achieves an accuracy of 97.62% on the 14-class task and 95.36% on the 28-class task of the SHREC’17 Track dataset, as well as 93.22% on the FPHA dataset.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"199 ","pages":"Article 105371"},"PeriodicalIF":5.2,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146070762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Household robot utilizing location information for human activity and habit understanding 利用位置信息来理解人类活动和习惯的家用机器人

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-27 DOI: 10.1016/j.robot.2026.105369

Tzu-Han Lin , Cih-An Chen , Chi-Hsiang Lo , Li-Chen Fu

This paper proposes an integrated system combining location estimation, human activity recognition (HAR), and plan recognition modules. In order to improve the HAR performance, we propose a location estimation system that fuses ResNet50-Places365 (Zhou et al., 2018) and our created estimator that leverages information on the distances between human and nearby objects. The location information from the location estimation system and the human skeleton information will be fed into HAR module governed by our developed activity-location graph convolutional neural network (AL-GCN). To explore more usage of the recognized activities, we propose a plan recognition system that updates the human’s plan knowledge base while taking into account human’s habits from time to time so as to make three important predictions, namely, next activity, objective, and plan. In our experiment, we evaluate our system on both dataset and real-world scenarios. In dataset evaluation, our location estimation system performs best with 92.83% accuracy, our AL-GCN model outperforms the state-of-the-art (SOTA) models with 94.33% accuracy on cross-subject evaluation, and our proposed plan recognition improves when habits are considered and knowledge base is updated. In the real-world experiments, the location estimation achieves 98% accuracy when in the living room, and our AL-GCN model improves the accuracy from 10% to 20% by including location information. Finally, our plan recognition shows that, by updating knowledge base, the predictions accuracy increases significantly.

本文提出了一种结合位置估计、人类活动识别（HAR）和计划识别模块的集成系统。为了提高HAR性能，我们提出了一种融合了ResNet50-Places365 （Zhou et al., 2018）和我们创建的估计器的位置估计系统，该估计器利用了人类与附近物体之间的距离信息。由我们开发的活动-位置图卷积神经网络（AL-GCN）控制的HAR模块将来自位置估计系统的位置信息和人体骨架信息馈送到HAR模块。为了探索识别活动的更多用途，我们提出了一种计划识别系统，它在不断更新人类的计划知识库的同时，考虑到人类的习惯，从而做出三个重要的预测，即下一个活动、目标和计划。在我们的实验中，我们在数据集和现实世界场景上评估了我们的系统。在数据集评估中，我们的位置估计系统以92.83%的准确率表现最佳，我们的AL-GCN模型在跨主题评估中以94.33%的准确率优于最先进的SOTA模型，当考虑习惯和知识库更新时，我们提出的计划识别得到改善。在现实世界的实验中，在客厅的位置估计准确率达到98%，我们的AL-GCN模型通过包含位置信息将准确率从10%提高到20%。最后，我们的计划识别表明，通过更新知识库，预测精度显著提高。

{"title":"Household robot utilizing location information for human activity and habit understanding","authors":"Tzu-Han Lin , Cih-An Chen , Chi-Hsiang Lo , Li-Chen Fu","doi":"10.1016/j.robot.2026.105369","DOIUrl":"10.1016/j.robot.2026.105369","url":null,"abstract":"<div><div>This paper proposes an integrated system combining location estimation, human activity recognition (HAR), and plan recognition modules. In order to improve the HAR performance, we propose a location estimation system that fuses ResNet50-Places365 (Zhou et al., 2018) and our created estimator that leverages information on the distances between human and nearby objects. The location information from the location estimation system and the human skeleton information will be fed into HAR module governed by our developed activity-location graph convolutional neural network (AL-GCN). To explore more usage of the recognized activities, we propose a plan recognition system that updates the human’s plan knowledge base while taking into account human’s habits from time to time so as to make three important predictions, namely, next activity, objective, and plan. In our experiment, we evaluate our system on both dataset and real-world scenarios. In dataset evaluation, our location estimation system performs best with 92.83% accuracy, our AL-GCN model outperforms the state-of-the-art (SOTA) models with 94.33% accuracy on cross-subject evaluation, and our proposed plan recognition improves when habits are considered and knowledge base is updated. In the real-world experiments, the location estimation achieves 98% accuracy when in the living room, and our AL-GCN model improves the accuracy from 10% to 20% by including location information. Finally, our plan recognition shows that, by updating knowledge base, the predictions accuracy increases significantly.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"199 ","pages":"Article 105369"},"PeriodicalIF":5.2,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146070763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Offshore Subsea IMR Operations: Review of the Automation Potential 海上水下IMR作业：自动化潜力综述

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-27 DOI: 10.1016/j.robot.2026.105365

Jannic Schurmann Larsen, Simon Pedersen, Jesper Liniger, Fredrik Fogh Sørensen

The demand for green energy continues to rise, with predictions indicating a 62-fold growth in floating wind. This offshore expansion, especially from wind, requires a corresponding increase in inspection, maintenance, and repair (IMR) operations. For offshore turbines, the marine segment accounts for half of fixed turbine O&M costs. One avenue to reduce these costs is through unmanned underwater vehicles (UUV) and automation. UUVs are becoming more specialized and automated, but most tasks remain only partially automated. It is estimated that the global energy cost in 2050 can be reduced by 0.5–0.6% from increased unmanned underwater vehicle automation, with wind contributing 0.4–0.5%. This study assesses automation levels and task frequencies across application domains, defined as types of offshore infrastructure, to calculate an automation priority score (APS), a weighted product of automation level and task frequency, indicating where economic and development benefits are. The automation level refers to a UUVs ability to operate without human intervention while performing tasks, regardless of industry use. The current highest APS is found in pipelines and jacket structures, with the lowest in wind turbines and dams. A sensitivity analysis evaluates the effects of increasing automation levels and anticipated annual growth. The study concludes that the wind sector will represent the highest automation priority in the future, providing the greatest economic incentive, especially for mooring lines. Visual inspection will increase due to AI, crawler-based UUVs will dominate circular structures IMR tasks, and autonomous underwater vehicle (AUVs) with subsea stations will handle frequent or long-range tasks.

对绿色能源的需求持续增长，预测表明浮动风的需求将增长62倍。这种海上扩张，特别是风能的扩张，需要相应增加检查、维护和维修（IMR）操作。对于海上涡轮机，海上部分占固定涡轮机运营成本的一半。降低这些成本的一个途径是通过无人水下航行器（UUV）和自动化。uuv正变得越来越专业化和自动化，但大多数任务仍然只是部分自动化。据估计，到2050年，无人潜航器自动化程度的提高可使全球能源成本降低0.5-0.6%，其中风能贡献0.4-0.5%。本研究评估了跨应用领域（定义为离岸基础设施类型）的自动化水平和任务频率，以计算自动化优先级评分（APS），这是自动化水平和任务频率的加权乘积，表明经济和发展效益在哪里。自动化级别指的是uuv在执行任务时无需人工干预的能力，无论行业用途如何。目前，管道和导管结构的APS最高，风力涡轮机和水坝的APS最低。敏感性分析评估了自动化水平提高和预期年增长的影响。该研究的结论是，风能行业将代表未来自动化的最高优先级，提供最大的经济激励，特别是对系泊线。由于人工智能，视觉检查将会增加，基于履带式的uuv将主导圆形结构的IMR任务，而带有海底站的自主水下航行器（auv）将处理频繁或远程任务。

{"title":"Offshore Subsea IMR Operations: Review of the Automation Potential","authors":"Jannic Schurmann Larsen, Simon Pedersen, Jesper Liniger, Fredrik Fogh Sørensen","doi":"10.1016/j.robot.2026.105365","DOIUrl":"10.1016/j.robot.2026.105365","url":null,"abstract":"<div><div>The demand for green energy continues to rise, with predictions indicating a 62-fold growth in floating wind. This offshore expansion, especially from wind, requires a corresponding increase in inspection, maintenance, and repair (IMR) operations. For offshore turbines, the marine segment accounts for half of fixed turbine O&M costs. One avenue to reduce these costs is through unmanned underwater vehicles (UUV) and automation. UUVs are becoming more specialized and automated, but most tasks remain only partially automated. It is estimated that the global energy cost in 2050 can be reduced by 0.5–0.6% from increased unmanned underwater vehicle automation, with wind contributing 0.4–0.5%. This study assesses automation levels and task frequencies across application domains, defined as types of offshore infrastructure, to calculate an automation priority score (APS), a weighted product of automation level and task frequency, indicating where economic and development benefits are. The automation level refers to a UUVs ability to operate without human intervention while performing tasks, regardless of industry use. The current highest APS is found in pipelines and jacket structures, with the lowest in wind turbines and dams. A sensitivity analysis evaluates the effects of increasing automation levels and anticipated annual growth. The study concludes that the wind sector will represent the highest automation priority in the future, providing the greatest economic incentive, especially for mooring lines. Visual inspection will increase due to AI, crawler-based UUVs will dominate circular structures IMR tasks, and autonomous underwater vehicle (AUVs) with subsea stations will handle frequent or long-range tasks.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"199 ","pages":"Article 105365"},"PeriodicalIF":5.2,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146070765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A review on vision-based control for multi-rotor aerial vehicles 多旋翼飞行器视觉控制研究进展

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-24 DOI: 10.1016/j.robot.2026.105366

Ana Maria P.S. Nascimento , Alisson V. Brito , Martin Saska , Tiago P. Nascimento

As the work speed and dynamics of multi-rotor aerial vehicles (MAVs) in real-world applications increase, vision-based control techniques have gained significant attention. By relying on visual input, environmental perception is enhanced, enabling tasks such as obstacle avoidance, landing, perching, and aerial manipulation. Thus, this article provides an overview of the latest advances in vision-based control for multi-rotor aerial robot systems, as well as future research directions that remain under exploration. We highlight a surge in research activity, particularly since 2010, driven by the proliferation of multi-rotor unmanned aerial systems. Additionally, we explore documented practical applications and identify the challenges that arise with this control paradigm. Finally, the primary aim of this survey is to provide a cohesive and accessible summary that contextualizes the mathematical foundations of vision-based control techniques applied to MAVs. This serves as a foundational resource for researchers venturing into the realm of visual-based control for MAVs. Ultimately, this work aims to provide a comprehensive review of recent advances in the field, directing readers to notable and successful works within the state-of-the-art landscape.

随着多旋翼飞行器（MAVs）在实际应用中的工作速度和动力学的提高，基于视觉的控制技术得到了广泛的关注。通过依赖视觉输入，环境感知得到增强，从而实现诸如避障、着陆、栖息和空中操纵等任务。因此，本文综述了多旋翼航空机器人系统视觉控制的最新进展，以及未来有待探索的研究方向。我们强调了研究活动的激增，特别是自2010年以来，受到多旋翼无人机系统扩散的推动。此外，我们还探讨了文档化的实际应用，并确定了这种控制范例所带来的挑战。最后，本调查的主要目的是提供一个有凝聚力和可访问的总结，将基于视觉的控制技术应用于自动驾驶汽车的数学基础语境化。这为研究人员冒险进入基于视觉的自动驾驶汽车控制领域提供了基础资源。最终，这项工作旨在提供该领域最新进展的全面回顾，指导读者在最先进的景观中值得注意和成功的作品。

{"title":"A review on vision-based control for multi-rotor aerial vehicles","authors":"Ana Maria P.S. Nascimento , Alisson V. Brito , Martin Saska , Tiago P. Nascimento","doi":"10.1016/j.robot.2026.105366","DOIUrl":"10.1016/j.robot.2026.105366","url":null,"abstract":"<div><div>As the work speed and dynamics of multi-rotor aerial vehicles (MAVs) in real-world applications increase, vision-based control techniques have gained significant attention. By relying on visual input, environmental perception is enhanced, enabling tasks such as obstacle avoidance, landing, perching, and aerial manipulation. Thus, this article provides an overview of the latest advances in vision-based control for multi-rotor aerial robot systems, as well as future research directions that remain under exploration. We highlight a surge in research activity, particularly since 2010, driven by the proliferation of multi-rotor unmanned aerial systems. Additionally, we explore documented practical applications and identify the challenges that arise with this control paradigm. Finally, the primary aim of this survey is to provide a cohesive and accessible summary that contextualizes the mathematical foundations of vision-based control techniques applied to MAVs. This serves as a foundational resource for researchers venturing into the realm of visual-based control for MAVs. Ultimately, this work aims to provide a comprehensive review of recent advances in the field, directing readers to notable and successful works within the state-of-the-art landscape.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"198 ","pages":"Article 105366"},"PeriodicalIF":5.2,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Human–human and human–robot co–manipulation: A biomechanical analysis of a joint carrying task 人-人和人-机器人协同操作：关节搬运任务的生物力学分析

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-24 DOI: 10.1016/j.robot.2026.105360

Fabian Goell , Bjoern Braunstein , Jule Heieis , Daniel Braun , Nadine Reißner , Kirill Safronov , Christian Weiser , Verena Schuengel , Kirsten Albracht

Assistive robots can support collaborative manipulation tasks such as carrying heavy or extended objects. As the human–human interaction is the basis for human–robot interaction, it is important to understand and quantify primarily haptic interaction. The subjects’ movements were recorded with a 3D motion capture system to determine spatio–temporal and upper and lower body kinematic parameters. The human–human interaction provided foundational data on human movement in collaborative manipulation tasks. The task with the robot revealed almost no changes in upper body kinematics, however, it was slower and showed adaptations of the human movement in the center of mass motion and in spatio–temporal parameters and lower body kinematics. This shows, that analyzing the interaction between humans and assistive robots focusing on human movement is essential for further developing assistive robots.

辅助机器人可以支持协作操作任务，例如搬运重物或扩展物体。由于人机交互是人机交互的基础，因此理解和量化触觉交互非常重要。通过三维运动捕捉系统记录受试者的运动，确定受试者的时空和上半身和下半身运动学参数。人与人之间的互动为协作操作任务中人类运动提供了基础数据。机器人在执行任务时上身运动学几乎没有变化，但速度较慢，在质心运动、时空参数和下体运动学方面表现出对人体运动的适应。这表明，分析人与辅助机器人之间的相互作用，关注人的运动是进一步开发辅助机器人的必要条件。

引用次数: 0

A half-cooperative strategy based Deep Q-network algorithm for multi-agent dynamic target hunting 基于半合作策略的深度q网络多智能体动态目标搜索算法

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-24 DOI: 10.1016/j.robot.2026.105368

Xiaoyan Wang, Xin Yu, Xi Fang

In fields like rescue, military, and law enforcement, effective target hunting is vital. Current algorithms for dynamic target hunting often fail in multi-target scenarios due to unknown environments and low cooperation efficiency. This paper introduces a multi-agent cooperative task planning model with a novel half-cooperative DQN algorithm for dynamic target hunting in unknown environments. The model randomly assigns positions to agents and targets, defines the action space for agents, and designs reward functions for real-world situations. The half-cooperative DQN algorithm uses prioritized experience replay for efficient learning and employs a half-cooperative strategy to enhance cooperation among agents, thereby improving hunting efficiency. Experimental results show that the half-cooperative DQN algorithm outperforms other improved DQN algorithms in terms of success rate, average boundary violations, and average time steps, highlighting its advantages and potential in dynamic target hunting.

在救援、军事和执法等领域，有效的目标搜寻至关重要。目前的动态目标搜索算法在多目标场景下，由于环境未知和协作效率低，往往会出现失败的情况。提出了一种基于半合作DQN算法的多智能体协作任务规划模型，用于未知环境下的动态目标搜索。该模型为智能体和目标随机分配位置，定义智能体的动作空间，并设计现实情况下的奖励函数。半合作DQN算法采用优先经验重播进行高效学习，采用半合作策略增强agent间的合作，从而提高搜索效率。实验结果表明，半合作DQN算法在成功率、平均边界违反和平均时间步长等方面都优于其他改进的DQN算法，突出了其在动态目标搜索方面的优势和潜力。

引用次数: 0

Adaptive artificial potential field method for small autonomous vehicles 小型自动驾驶汽车自适应人工势场法

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-23 DOI: 10.1016/j.robot.2026.105364

Kemal Ihsan Kilic, Aurélien Desoeuvres, Casper Bak Pedersen, Alex Elkjær Vasegaard, Peter Nielsen

Artificial Potential Field (APF) is one of the path planning and obstacle avoidance methods used for its simplicity and effectiveness. The goal’s attractive force and the obstacles’ repulsive forces are modeled and considered to act upon the vehicle or robot. However, in practice, the classical APF faces challenges like local minima. We propose to study several enhancements and their combinations in order to show how they can create even more efficient algorithms to overcome these challenges. Those enhancements include tangential force, inertia-inspired force, and a local minima detection (l.m.d.) and reaction scheme by adding virtual obstacles and dynamically changing coefficients. All of them keep the methods as lightweight as possible while improving the classical APF. The tangential force provides smoother paths and avoids local minima cases. The dynamic change of APF’s parameters, coupled with the addition of virtual obstacles when detecting local minima, provides an efficient way to escape them. The inertia-inspired force can be used to smooth the trajectory when only obstacles in front of the vehicle are taken into account. We defined performance metrics to assess the path completion, path quality, and processing time to compare the proposed enhancements with the base case of classical APF. We benchmarked the proposed methods in different environments for a holonomic robot and a simplified bicycle. The proposed adaptive APF with inertial force extension completed 87.5% of the tests while the classical APF completed only 43.8% of them. On the other side, tangential versions of APF reduce the path length deviation by 8% and the curvature by 20% in simple cases. The code is available on: https://github.com/Glawal/APFproject/tree/paper1.

人工势场（Artificial Potential Field， APF）是一种简单有效的路径规划和避障方法。对目标的吸引力和障碍物的排斥力进行建模，并考虑它们对车辆或机器人的作用。然而，在实际应用中，经典的有源滤波器面临着局部最小值等挑战。我们建议研究几种增强及其组合，以展示它们如何创建更有效的算法来克服这些挑战。这些改进包括切向力，惯性激励力，以及通过添加虚拟障碍物和动态变化系数的局部最小检测（l.m.d.）和反应方案。所有这些方法都在改进经典APF的同时尽可能地保持轻量级。切向力提供了更平滑的路径，避免了局部极小的情况。APF参数的动态变化，加上在检测局部极小值时添加虚拟障碍物，提供了一种有效的逃避方法。在只考虑前方障碍物的情况下，利用惯性激励可以使飞行器的轨迹平滑。我们定义了性能指标来评估路径完成、路径质量和处理时间，以将所提出的增强与经典APF的基本情况进行比较。我们在一个完整机器人和一个简化的自行车的不同环境中对所提出的方法进行了基准测试。提出的惯性力扩展自适应APF完成了87.5%的测试，而经典APF仅完成了43.8%的测试。另一方面，切向版本的APF在简单情况下减少了8%的路径长度偏差和20%的曲率。代码可在https://github.com/Glawal/APFproject/tree/paper1上获得。

{"title":"Adaptive artificial potential field method for small autonomous vehicles","authors":"Kemal Ihsan Kilic, Aurélien Desoeuvres, Casper Bak Pedersen, Alex Elkjær Vasegaard, Peter Nielsen","doi":"10.1016/j.robot.2026.105364","DOIUrl":"10.1016/j.robot.2026.105364","url":null,"abstract":"<div><div>Artificial Potential Field (APF) is one of the path planning and obstacle avoidance methods used for its simplicity and effectiveness. The goal’s attractive force and the obstacles’ repulsive forces are modeled and considered to act upon the vehicle or robot. However, in practice, the classical APF faces challenges like local minima. We propose to study several enhancements and their combinations in order to show how they can create even more efficient algorithms to overcome these challenges. Those enhancements include tangential force, inertia-inspired force, and a local minima detection (l.m.d.) and reaction scheme by adding virtual obstacles and dynamically changing coefficients. All of them keep the methods as lightweight as possible while improving the classical APF. The tangential force provides smoother paths and avoids local minima cases. The dynamic change of APF’s parameters, coupled with the addition of virtual obstacles when detecting local minima, provides an efficient way to escape them. The inertia-inspired force can be used to smooth the trajectory when only obstacles in front of the vehicle are taken into account. We defined performance metrics to assess the path completion, path quality, and processing time to compare the proposed enhancements with the base case of classical APF. We benchmarked the proposed methods in different environments for a holonomic robot and a simplified bicycle. The proposed adaptive APF with inertial force extension completed 87.5% of the tests while the classical APF completed only 43.8% of them. On the other side, tangential versions of APF reduce the path length deviation by 8% and the curvature by 20% in simple cases. The code is available on: <span><span>https://github.com/Glawal/APFproject/tree/paper1</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"198 ","pages":"Article 105364"},"PeriodicalIF":5.2,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Towards robot affective appraisal linking inner speech and emotion 面向机器人情感评价，将内在言语与情感联系起来

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-21 DOI: 10.1016/j.robot.2026.105363

Arianna Pipitone , Sophia Corvaia , Antonio Chella

Recent studies in Robotics and AI suggest that robots “thinking out loud” can foster positive human feedback and support collaborative goal achievement. By externalizing their internal reasoning, robots enhance transparency and explainability, which are crucial for trust and robustness in human–robot interaction.

This work investigates the role of robot inner speech in supporting affective appraisal, focusing on the emergence, coherence, and interpretability of emotionally grounded evaluations. While the relationship between inner speech and emotion has been extensively explored within human psychology and cognitive science, this work is the first to operationalize this connection within a robotic architecture for affective appraisal.

Grounded in appraisal theories, the proposed model employs inner speech to simulate internal reflection, enabling the identification and evaluation of contextual variables relevant to affective assessment. Through this internal dialogue, the robot structures its appraisal process and externalizes it, allowing human partners to access and interpret the underlying affective reasoning.

The model is evaluated by comparing its appraisal dynamics with normative emotional patterns observed in adults under stress, and by assessing the interpretability of the robot’s affective behavior through human observation. Results indicate that the model produces coherent and context-sensitive evaluations, improving upon a widely adopted computational model of emotion in terms of plausibility and transparency.

最近在机器人和人工智能领域的研究表明，机器人“大声思考”可以促进积极的人类反馈，并支持协作目标的实现。通过将其内部推理外部化，机器人提高了透明度和可解释性，这对于人机交互中的信任和鲁棒性至关重要。这项工作调查了机器人内部语言在支持情感评估中的作用，重点关注情感基础评估的出现、一致性和可解释性。虽然内在言语和情感之间的关系在人类心理学和认知科学中得到了广泛的探索，但这项工作是第一次将这种联系应用于情感评估的机器人架构中。该模型以评估理论为基础，采用内部言语模拟内部反思，从而识别和评估与情感评估相关的上下文变量。通过这种内部对话，机器人构建其评估过程并将其外部化，允许人类伴侣访问并解释潜在的情感推理。该模型通过将其评估动态与成人在压力下观察到的规范情绪模式进行比较，并通过人类观察评估机器人情感行为的可解释性来评估。结果表明，该模型产生连贯和上下文敏感的评估，在合理性和透明度方面改进了广泛采用的情感计算模型。

{"title":"Towards robot affective appraisal linking inner speech and emotion","authors":"Arianna Pipitone , Sophia Corvaia , Antonio Chella","doi":"10.1016/j.robot.2026.105363","DOIUrl":"10.1016/j.robot.2026.105363","url":null,"abstract":"<div><div>Recent studies in Robotics and AI suggest that robots “thinking out loud” can foster positive human feedback and support collaborative goal achievement. By externalizing their internal reasoning, robots enhance transparency and explainability, which are crucial for trust and robustness in human–robot interaction.</div><div>This work investigates the role of robot inner speech in supporting affective appraisal, focusing on the emergence, coherence, and interpretability of emotionally grounded evaluations. While the relationship between inner speech and emotion has been extensively explored within human psychology and cognitive science, this work is the first to operationalize this connection within a robotic architecture for affective appraisal.</div><div>Grounded in appraisal theories, the proposed model employs inner speech to simulate internal reflection, enabling the identification and evaluation of contextual variables relevant to affective assessment. Through this internal dialogue, the robot structures its appraisal process and externalizes it, allowing human partners to access and interpret the underlying affective reasoning.</div><div>The model is evaluated by comparing its appraisal dynamics with normative emotional patterns observed in adults under stress, and by assessing the interpretability of the robot’s affective behavior through human observation. Results indicate that the model produces coherent and context-sensitive evaluations, improving upon a widely adopted computational model of emotion in terms of plausibility and transparency.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"198 ","pages":"Article 105363"},"PeriodicalIF":5.2,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Efficient distributed multi-robot SLAM under missed detections and clutter with labeled multi-Bernoulli map 基于标记多伯努利地图的缺席检测和杂波下的高效分布式多机器人SLAM

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-20 DOI: 10.1016/j.robot.2026.105358

Lin Gao , Giorgio Battistelli , Luigi Chisci

The labeled multi-Bernoulli (LMB) propagation of the environmental map has demonstrated its effectiveness for single-vehicle simultaneous localization and mapping (SLAM), yielding the LMB-SLAM method. In applications of LMB-SLAM, the robot is equipped with sensors (such as radar) that provide detection points of landmarks, so that the main challenge of performing SLAM is to tackle data association between landmarks and measurements under missed detections and clutter. This paper provides two contributions on LMB-SLAM. First, we improve its computational efficiency by an appropriate design of the proposal distribution that approximates the posterior of the vehicle pose. In this way, particles can be sampled in a more efficient way thus remarkably reducing their number, and the consequent computational load required to achieve a given performance level. Secondly, we extend LMB-SLAM to the multi-vehicle case, assuming that relative initial poses among vehicles are known, by suitable design of a map fusion method that allows to generate a more accurate and complete map in both explored and unexplored regions so as to improve both localization and mapping performance by vehicle cooperation. The improved performance of the proposed, single-vehicle and multi-vehicle, LMB-SLAM algorithms is assessed via experiments on both simulated and real data.

环境地图的标记多伯努利（LMB）传播证明了其在单车辆同时定位和地图绘制（SLAM）中的有效性，从而产生了LMB-SLAM方法。在LMB-SLAM的应用中，机器人配备了传感器（如雷达）来提供地标的检测点，因此执行SLAM的主要挑战是在漏检和杂波情况下处理地标与测量之间的数据关联。本文对LMB-SLAM有两方面的贡献。首先，我们通过适当设计近似于车辆姿态后验的建议分布来提高其计算效率。通过这种方式，粒子可以以更有效的方式采样，从而显着减少它们的数量，以及达到给定性能水平所需的随之而来的计算负载。其次，我们将LMB-SLAM扩展到多车情况下，假设车辆之间的相对初始姿态已知，通过设计合适的地图融合方法，在探索区域和未探索区域生成更精确和完整的地图，从而通过车辆合作提高定位和地图绘制性能。通过模拟和真实数据的实验，对所提出的单载和多载LMB-SLAM算法的改进性能进行了评估。

{"title":"Efficient distributed multi-robot SLAM under missed detections and clutter with labeled multi-Bernoulli map","authors":"Lin Gao , Giorgio Battistelli , Luigi Chisci","doi":"10.1016/j.robot.2026.105358","DOIUrl":"10.1016/j.robot.2026.105358","url":null,"abstract":"<div><div>The <em>labeled multi-Bernoulli</em> (LMB) propagation of the environmental map has demonstrated its effectiveness for single-vehicle <em>simultaneous localization and mapping</em> (SLAM), yielding the LMB-SLAM method. In applications of LMB-SLAM, the robot is equipped with sensors (such as radar) that provide detection points of landmarks, so that the main challenge of performing SLAM is to tackle data association between landmarks and measurements under missed detections and clutter. This paper provides two contributions on LMB-SLAM. First, we improve its computational efficiency by an appropriate design of the proposal distribution that approximates the posterior of the vehicle pose. In this way, particles can be sampled in a more efficient way thus remarkably reducing their number, and the consequent computational load required to achieve a given performance level. Secondly, we extend LMB-SLAM to the multi-vehicle case, assuming that relative initial poses among vehicles are known, by suitable design of a map fusion method that allows to generate a more accurate and complete map in both explored and unexplored regions so as to improve both localization and mapping performance by vehicle cooperation. The improved performance of the proposed, single-vehicle and multi-vehicle, LMB-SLAM algorithms is assessed via experiments on both simulated and real data.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"198 ","pages":"Article 105358"},"PeriodicalIF":5.2,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037685","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modeling and solving UAV 3D trajectory planning using an efficient multi-mechanism enhanced chaos game optimization algorithm 基于高效多机构增强型混沌博弈优化算法的无人机三维轨迹规划建模与求解

IF 5.2 2区计算机科学 Q1 AUTOMATION & CONTROL SYSTEMS

Robotics and Autonomous Systems

Pub Date : 2026-01-20 DOI: 10.1016/j.robot.2026.105361

Jingsen Liu , Fei Li , Yu Li , Huan Zhou

Three-dimensional (3D) trajectory planning for unmanned aerial vehicles (UAVs) is one of the critical technologies for achieving Self-guided flying, targeting computing a safe, efficient, and energy-optimal flight path. To satisfy the realistic flight needs of unmanned aerial vehicles, this paper establishes a mathematical model for 3D trajectory planning with optimization objectives including flight energy consumption, path length, flight altitude, flight threats, and trajectory smoothness. Simultaneously, the spherical coordinate system is utilized to optimize the UAV’s flight attitude and path, and cubic quasi-uniform B-spline curves are employed to smooth the flight trajectory. To better address and optimize the UAV 3D trajectory planning problem, a boosted chaos game optimization algorithm (SACGO) is put forward in this research, which effectively equilibrates the algorithm’s worldwide and localized search functionalities through a phased multi-mechanism mutation strategy. Additionally, an adaptive convergence factor based on dual random perturbations and cyclic boundary handling is introduced, enhancing the quality of candidate solutions while accelerating convergence speed. Regarding the analysis of theoretical algorithms, the improved algorithm has the exact same complexity in time as the original CGO, according to a time complexity study. Furthermore, probability measure methods are used to prove the global convergence of SACGO. For algorithm performance testing, we first performed a coupled sensitivity analysis on the two perturbation parameters within the improvement mechanism to identify their optimal values. Subsequently, SACGO is contrasted with 7 representative high-performance algorithms across multiple dimensions and evaluation methods on the CEC2022 and CEC2017 benchmark suites. The findings show that SACGO's convergence speed, solution correctness, and stability are all markedly improved by the suggested modifications. Finally, SACGO is used for UAV trajectory planning and tested against other algorithms in six scenarios with varying complexities. The solving results of 8 algorithms, including SACGO, are shown. SACGO efficiently plans flight paths under all constraints, comprehensively optimizes all objectives, and generates trajectories that are not only safe and feasible but also achieve the lowest trajectory cost.

无人机的三维轨迹规划是实现无人机自制导飞行的关键技术之一，其目标是计算安全、高效、能量最优的飞行路径。为满足无人机实际飞行需求，建立了以飞行能耗、路径长度、飞行高度、飞行威胁、轨迹平整度为优化目标的无人机三维轨迹规划数学模型。同时，利用球坐标系对无人机的飞行姿态和轨迹进行优化，利用三次准均匀b样条曲线对飞行轨迹进行平滑。为了更好地解决和优化无人机三维轨迹规划问题，提出了一种改进的混沌博弈优化算法（SACGO），该算法通过阶段性多机制突变策略有效地平衡了算法的全局搜索和局部搜索功能。此外，引入了基于对偶随机扰动和循环边界处理的自适应收敛因子，提高了候选解的质量，加快了收敛速度。在理论算法分析方面，根据时间复杂度研究，改进算法在时间复杂度上与原CGO完全相同。在此基础上，利用概率测度方法证明了SACGO的全局收敛性。为了测试算法的性能，我们首先对改进机制内的两个扰动参数进行了耦合灵敏度分析，以确定其最优值。随后，在CEC2022和CEC2017基准套件上，将SACGO与7种具有代表性的跨多维高性能算法和评估方法进行对比。结果表明，改进后的SACGO算法的收敛速度、解的正确性和稳定性均有显著提高。最后，SACGO用于无人机轨迹规划，并在六种不同复杂情况下与其他算法进行测试。给出了包括SACGO在内的8种算法的求解结果。SACGO能够高效地规划各种约束条件下的飞行路径，对所有目标进行综合优化，生成安全可行且轨迹成本最低的轨迹。

{"title":"Modeling and solving UAV 3D trajectory planning using an efficient multi-mechanism enhanced chaos game optimization algorithm","authors":"Jingsen Liu , Fei Li , Yu Li , Huan Zhou","doi":"10.1016/j.robot.2026.105361","DOIUrl":"10.1016/j.robot.2026.105361","url":null,"abstract":"<div><div>Three-dimensional (3D) trajectory planning for unmanned aerial vehicles (UAVs) is one of the critical technologies for achieving Self-guided flying, targeting computing a safe, efficient, and energy-optimal flight path. To satisfy the realistic flight needs of unmanned aerial vehicles, this paper establishes a mathematical model for 3D trajectory planning with optimization objectives including flight energy consumption, path length, flight altitude, flight threats, and trajectory smoothness. Simultaneously, the spherical coordinate system is utilized to optimize the UAV’s flight attitude and path, and cubic quasi-uniform B-spline curves are employed to smooth the flight trajectory. To better address and optimize the UAV 3D trajectory planning problem, a boosted chaos game optimization algorithm (SACGO) is put forward in this research, which effectively equilibrates the algorithm’s worldwide and localized search functionalities through a phased multi-mechanism mutation strategy. Additionally, an adaptive convergence factor based on dual random perturbations and cyclic boundary handling is introduced, enhancing the quality of candidate solutions while accelerating convergence speed. Regarding the analysis of theoretical algorithms, the improved algorithm has the exact same complexity in time as the original CGO, according to a time complexity study. Furthermore, probability measure methods are used to prove the global convergence of SACGO. For algorithm performance testing, we first performed a coupled sensitivity analysis on the two perturbation parameters within the improvement mechanism to identify their optimal values. Subsequently, SACGO is contrasted with 7 representative high-performance algorithms across multiple dimensions and evaluation methods on the CEC2022 and CEC2017 benchmark suites. The findings show that SACGO's convergence speed, solution correctness, and stability are all markedly improved by the suggested modifications. Finally, SACGO is used for UAV trajectory planning and tested against other algorithms in six scenarios with varying complexities. The solving results of 8 algorithms, including SACGO, are shown. SACGO efficiently plans flight paths under all constraints, comprehensively optimizes all objectives, and generates trajectories that are not only safe and feasible but also achieve the lowest trajectory cost.</div></div>","PeriodicalId":49592,"journal":{"name":"Robotics and Autonomous Systems","volume":"198 ","pages":"Article 105361"},"PeriodicalIF":5.2,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0