Artificial Life and Robotics最新文献

{"title":"Application of lightweight 3D mouth localization in vision-guided assistive feeding systems","authors":"Fei Liu,&nbsp;Zhi Li,&nbsp;Chilan Cai","doi":"10.1007/s10015-025-01085-3","DOIUrl":"10.1007/s10015-025-01085-3","url":null,"abstract":"<div><p>Assistive feeding robots aim to improve the quality of life for individuals with physical disabilities, yet existing solutions often suffer from bulky perception modules and low system responsiveness. Accurate and lightweight 3D mouth localization is essential to enable real-time, user-aware feeding interaction. To develop a lightweight and efficient vision-based system for real-time 3D mouth detection and adaptive assistive feeding. A pruned YOLOv5n model was employed for efficient and accurate mouth detection. Using an RGB-D camera and hand–eye calibration, 2D detection results were converted into 3D mouth coordinates within the robot’s base frame. A rule-based control strategy was adopted: feeding is initiated if the mouth remains open for more than 3 s and terminated if the mouth is not detected for 3 s. The feeding process consists of a pre-demonstrated scooping trajectory, followed by smooth motion toward the estimated mouth center. The proposed system achieves real-time performance with a significantly reduced model size (2.86 MB), parameter count (1.33 M), and computation cost (3.1 GFLOPs), compared to the original YOLOv5n baseline (3.65 MB, 1.77 M, 4.2 GFLOPs). Despite this compression, the pruned model maintains high detection accuracy with mAP0.5 of 98.8% and mAP0.5: 0.95 of 61.3%. Experimental evaluations confirm robust 3D mouth localization and stable feeding execution under user head movements. The model runs at approximately 16.2 FPS on an Intel N150 CPU, validating its suitability for real-time deployment on low-power edge devices. The control logic enables natural and safe human–robot interaction during assistive feeding.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 2","pages":"611 - 621"},"PeriodicalIF":0.8,"publicationDate":"2025-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147938759","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Group dynamics of Plecoglossus altivelis under different light environments","authors":"Kohei Ohashi,&nbsp;Shoma Kamata,&nbsp;Rei Hiraoka,&nbsp;Hiroaki Kawashima,&nbsp;Hitoshi Habe,&nbsp;Takayuki Niizato","doi":"10.1007/s10015-025-01080-8","DOIUrl":"10.1007/s10015-025-01080-8","url":null,"abstract":"<div><p>Collective behavior in nature exemplifies self-organization through local interactions, with fish schooling being a compelling case. While illuminance affects both individual physiological traits and group dynamics, light wavelength has been shown to influence individual physiological traits, such as growth and endocrine secretion. However, its role in shaping collective behavior remains unclear. In this study, we used <i>Plecoglossus altivelis</i> to investigate schooling behavior under three light conditions: darkness, blue light, and red light. The key findings included: (1) fish formed fewer groups in darkness, with larger inter-individual distances and lower orientation alignment and (2) under blue light, fish exhibited tighter, more aligned schools than the fluid and variable group formations under red light. These results suggest that light wavelengths significantly impact schooling behavior and have potential applications in aquaculture and fish guidance systems.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 1","pages":"63 - 68"},"PeriodicalIF":0.8,"publicationDate":"2025-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147342497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Metamorphic transitions in Lenia through Dynamic Growth Mechanisms","authors":"Yoshihiko Kayama","doi":"10.1007/s10015-025-01081-7","DOIUrl":"10.1007/s10015-025-01081-7","url":null,"abstract":"<div><p>This study investigates dynamic transitions within the Lenia framework, a continuous cellular automaton that enables complex, autonomous lifeforms. We introduce Dynamic Growth Mechanisms (DGM), a novel approach consisting of three extensions: Dual-phase Growth, Kernel Scaling, and Kernel Translation. These mechanisms aim to promote self-organization and metamorphosis-like transitions from simple initial configurations. Dual-phase Growth combines Original and Asymptotic Lenia’ s growth rules. Kernel Scaling treats the kernel as an evolving cell attribute. Crucially, Kernel Translation breaks rotational symmetry by applying an offset vector (<span>(varvec{{delta }})</span>) to the kernel. Simulations of Dual-phase Lenia and DGM Lenia reveal complex behaviors, including pair production, fusion, and transitions reminiscent of insect metamorphosis. These dynamic behaviors and lifeform transitions are further illustrated in a supplementary online video gallery available at https://y-kayama.github.io/dgm-lenia/. DGM Lenia exhibits a variety of stable lifeforms and their transitions, suggesting potential evolutionary pathways from simple to more complex structures. Significantly, our detailed analysis reveals that the offset vector (<span>(varvec{{delta }})</span>) in Kernel Translation plays a critical role in shaping the parameter space for organism emergence and transitions. Specifically, appropriate <span>(varvec{{delta }})</span> values establish a distinct region between areas dominated by Original Lenia and Asymptotic Lenia growth rules, significantly enhancing the emergence of diverse lifeforms, particularly Optium/Dioptium-class organisms and their associated transitions. Our findings highlight the potential of dynamic transitions between lifeforms and underscore the crucial role of symmetry breaking for promoting biodiversity, providing new avenues for exploring evolutionary mechanisms and stability in artificial life.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 2","pages":"562 - 574"},"PeriodicalIF":0.8,"publicationDate":"2025-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147938387","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolutionary design of group organization and controllers in heterogeneous robotic swarms","authors":"Asad Razzaq,&nbsp;Toshiyuki Yasuda","doi":"10.1007/s10015-025-01083-5","DOIUrl":"10.1007/s10015-025-01083-5","url":null,"abstract":"<div><p>Swarm robotics systems (SRS) enable robots to coordinate like social organisms, achieving complex tasks through collective behavior that exceeds the capabilities of individual robots. This study proposes an evolutionary approach for the automatic design of heterogeneous controllers and sub-team compositions in robotic swarms performing a cooperative transport task, in contrast to conventional homogeneous SRS<span>(_s)</span>. The proposed method evolves both controller parameters and their distribution across the swarm. Utilizing artificial neural networks and evolutionary strategies, the system dynamically adapts robot roles according to task complexity. Experimental results demonstrate improvements in performance over conventional approaches, highlighting the benefits of adaptive heterogeneity in swarm robotics.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 1","pages":"294 - 301"},"PeriodicalIF":0.8,"publicationDate":"2025-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147341081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clustering analysis of long-term robot motion patterns generated by cat motion trained gMLP","authors":"Koki Sato,&nbsp;Ryo Kobayashi,&nbsp;Sho Yamauchi,&nbsp;Keiji Suzuki,&nbsp;Sho’ji Suzuki","doi":"10.1007/s10015-025-01077-3","DOIUrl":"10.1007/s10015-025-01077-3","url":null,"abstract":"<div><p>A foundational framework for diverse robot motion generation is proposed and its effectiveness in creating appealing robots is verified in this paper. The framework uses a two-step approach: train a large language model (LLM) and use it to control the robot’s behavior. During the learning phase, the training dataset is generated by extracting 2D keypoints from cat videos, converting these 2D keypoints into 3D keypoints, and calculating joint angles using inverse kinematics based on joint positions. These data, comprising joints and their positions, were transformed into a language-like format, termed “motion language.” These converted data are then used to train the gMLP, a large language model. The gMLP showed its lowest validation loss after 2 epochs, so the model trained up to this point is used for motion generation for over 30 min. These motion data generated by trained gMLP were analyzed by clustering method. The results showed that the analysis of 5-s intervals revealed five distinct movement patterns. Similarly, in 10-s intervals, three distinct movement patterns were observed.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 1","pages":"159 - 167"},"PeriodicalIF":0.8,"publicationDate":"2025-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147341080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimation of stress-responsive hemodynamics from near-infrared facial image using t-SNE","authors":"Shonosuke Ohyama,&nbsp;Kent Nagumo,&nbsp;Akio Nozawa","doi":"10.1007/s10015-025-01076-4","DOIUrl":"10.1007/s10015-025-01076-4","url":null,"abstract":"<div><p>When humans are exposed to external stimuli, they exhibit a stress response. This response can be categorized into three types based on coping mechanisms: active coping styles, passive coping styles, and no coping. These coping styles can be differentiated by characteristic variation patterns in hemodynamic indices, such as mean blood pressure (MBP), cardiac output (CO), and total peripheral resistance (TPR). Current hemodynamic measurement methods are contact-based and time-consuming, highlighting the need for a non-contact technique to quickly assess stress-coping styles. In this study, we focused on facial blood flow information captured in the near-infrared spectrum and aimed to estimate stress coping styles by applying dimensionality reduction and sparse modeling to near-infrared facial image (NIFI). As an initial step, dimensionality reduction was applied to NIFI to visualize divergence according to stress-coping styles in a two-dimensional space. Subsequently, k-nearest neighbor (k-NN) labeling and sparse modeling were employed to classify stress-coping styles. The proposed model achieved a classification accuracy of 70.5<span>(%)</span>. However, in some cases, insufficient separation between coping styles in the two-dimensional space contributed to limited generalization performance. Future work will focus on improving generalization by refining the dimensionality reduction and evaluation methods, and by addressing physiological delays inherent in blood flow responses. This study employs NIFI, which is distinct from Near-Infrared Spectroscopy (NIRS). NIFI assesses autonomic responses through facial skin blood flow and is not intended as a measure of brain activity.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 2","pages":"504 - 514"},"PeriodicalIF":0.8,"publicationDate":"2025-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147938380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Numerical evaluation of a mobile underwater manipulator with a structure–wall suction mechanism","authors":"Norimitsu Sakagami,&nbsp;Makoto Iwasaki,&nbsp;Masatoshi Fukami,&nbsp;Yuki Tanaka,&nbsp;Aoi Koshioka,&nbsp;Atsushi Kakogawa","doi":"10.1007/s10015-025-01073-7","DOIUrl":"10.1007/s10015-025-01073-7","url":null,"abstract":"<div><p>This paper proposes a mobile underwater robotic manipulator equipped with a suction mechanism consisting of a marine thruster and a disk plate to ensure its position-keeping performance. The marine thruster induces water flow between the disk plate and a structure’s surface. The water flow produces negative pressure to maintain stable contact between the manipulator and the structure’s surface, thereby ensuring reliable manipulation. Even when hydrodynamic forces act on the manipulator, the suction force keeps it stable on the structure’s surface. For this work, a prototype underwater manipulator with a suction mechanism was designed and developed. Based on measurements of the force and moment generated by the suction mechanism, the stabilization performance of the manipulator was evaluated numerically when a disturbance water flow is present or when the manipulator moves dynamically. Results confirmed that the prototype manipulator can maintain its position and orientation even in a disturbance water flow and during dynamic motions of the manipulator.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"30 4","pages":"812 - 819"},"PeriodicalIF":0.8,"publicationDate":"2025-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10015-025-01073-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529669","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Route planning of an unmanned aerial vehicle using ant colony optimization for environment including a no-fly zone and obstacles","authors":"Yoshiaki Katada,&nbsp;Takumi Kato,&nbsp;Nao Nakajima","doi":"10.1007/s10015-025-01079-1","DOIUrl":"10.1007/s10015-025-01079-1","url":null,"abstract":"<div><p>The shortest route should be desired when considering flight time or battery capacity for an unmanned aerial vehicle (UAV) operation. Flight areas can include no-fly zones or obstacles that a UAV can avoid or fly over. This study addresses UAV route planning considering such no-fly zones and obstacles. We planned the routes of a UAV avoiding no-fly zones and flying over obstacles by ant colony optimization in computer simulations, then conducted the automated flight experiment on a real UAV based on the obtained routes, and confirmed the effectiveness of this approach. We show that the ant colony optimization devising program establishment can make possible UAV route planning for flight areas with no-fly zones and obstacles.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 1","pages":"254 - 264"},"PeriodicalIF":0.8,"publicationDate":"2025-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147340792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Target-oriented exploration guided by anticipated returns","authors":"Akane Tsuboya,&nbsp;Yu Kono,&nbsp;Tatsuji Takahashi","doi":"10.1007/s10015-025-01068-4","DOIUrl":"10.1007/s10015-025-01068-4","url":null,"abstract":"<div><p>The objective of a reinforcement learning agent is to discover the best actions through exploration. However, standard methods such as <span>(epsilon)</span>-greedy and intrinsic motivation may rely excessively on random selection or the pursuit of novelty. As a result, they can drive exploration toward states unrelated to the actual learning objective, potentially leading to inefficient learning and delayed convergence. We propose a novel deep reinforcement learning method that prioritizes achieving a target return over maximizing return. This method directly utilizes the target return as a guiding signal for exploration, prioritizing actions that are more likely to achieve the target return. In addition, when the target has not yet been achieved, our agent actively explores the less frequently selected actions to discover better ones. Through experiments on a motion control task and a navigation task, our method demonstrated more stable and robust performance than standard methods, achieving higher returns with fewer episodes. These findings suggest that exploration based on target return can be an effective approach for practical applications whose priority is achieving a specific performance level under limited resources.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"30 4","pages":"752 - 763"},"PeriodicalIF":0.8,"publicationDate":"2025-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145529537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reducing collision counts at intersections using reinforcement learning-based adaptive speed control","authors":"Kaito Kumagae,&nbsp;Tenta Suzuki,&nbsp;Mao Tobisawa,&nbsp;Tomohiro Harada,&nbsp;Johei Matsuoka,&nbsp;Yuki Itoh,&nbsp;Clive Jancen Kawaoto,&nbsp;Kiyohiko Hattori","doi":"10.1007/s10015-025-01071-9","DOIUrl":"10.1007/s10015-025-01071-9","url":null,"abstract":"<div><p>Research on cooperative control of autonomous vehicles using reinforcement learning has gained attention in the recent years. Conventional intersection control using traffic signals faces challenges in flexibly responding to changing traffic conditions, making balancing safety improvement and traffic flow optimization difficult. This study proposes an intersection agent that dynamically adjusts the entry speed of vehicles at unsignalized intersections and explores a method to acquire control policies through reinforcement learning. In the proposed method, the intersection agent observes the number of vehicles from each entry direction and implements continuous speed limits for vehicles in each direction, enabling control that responds to congestion conditions and collision risks. The Proximal Policy Optimization (PPO) deep reinforcement learning algorithm was used to learn the control policy. Simulation evaluation results showed that compared with Matsuda et al.'s method (Kishi et al. in 27:513–520, 2022), the average number of vehicles passing through was 496.4 for the proposed method and 488.6 for the conventional method, confirming that control efficiency was maintained. In addition, the number of collisions within the intersection was 9.8 on average for the proposed method, compared to 31.3 for the conventional method, achieving a significant reduction. In a four-way intersection scenario, the proposed method achieved an average vehicle throughput of 218.6 and recorded 3.3 collisions, while the baseline method yielded 181.3 and 11.0, respectively. This represents a 20.6% increase in throughput and a 70% reduction in collisions, demonstrating a significant improvement in both traffic efficiency and safety. The proposed method learned an adaptive control policy that does not impose speed limits when traffic conditions are stable, but temporarily strengthens restrictions only during congestion or high-collision risk situations, showing potential for simultaneously optimizing traffic flow and improving safety.</p></div>","PeriodicalId":46050,"journal":{"name":"Artificial Life and Robotics","volume":"31 1","pages":"228 - 238"},"PeriodicalIF":0.8,"publicationDate":"2025-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147339939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}