地球表面 71% 被海洋覆盖，这片孕育生命的蓝色疆域不仅蕴藏着海量自然资源，更是调节全球生态的关键枢 纽。然而，超过 90% 的海洋区域仍处于人类探索的 "盲区"，加之人类活动对海洋环境的影响日益加深，对 深海进行低干扰、长时程、高适应性的监测，已成为海洋科学与生态保护的迫切需求。 研究团队提出了一种全新的驱动 /变形系统，核心是一种称为"后屈曲缺口板"（PBNP）的创新结构。 这种 设计模仿了蝠鲼的胸鳍，通过巧妙的机械结构将微小的线性运动转化为大幅度的鳍片拍动。 传统水下探测设备却难以突破环境局限：依赖螺旋桨等刚性部件的机器人，虽推力强劲，却在复杂流场、 狭 窄空间 或脆弱生态区 "水土不服"，容易卡在礁石缝隙中。而仿生机器鱼凭借柔软可变形的身体和仿生泳姿展 现出优势，仍面临游动模式单一、深海极端温度及非结构化环境适应性差等难题。机器鱼既能够在狭窄空间稳 定穿行、又能够在开阔水域欢快地游动，是 仿生机器鱼设计的核心挑战 。 而浙江大学研发的这款仿生机器鱼，恰恰打破了这一困境。 它跳出传统水下探测装置的设计框架，通过创新 的驱动 /变形系统，为复杂海洋环境下的低干扰探测与生态监测，开辟了全新路径。 这项研究， ...

Group 1 - The Hangzhou Cultural Expo showcases the integration of technology and culture, featuring AI-generated content and immersive experiences [2][3] - The event introduces a new "Digital New City" pavilion, highlighting cutting-edge technologies such as AI-generated art and embodied robotics [2][3] - The "Star Gathering Future" exhibition celebrates the 10th anniversary of the Hangzhou Cultural Innovation Fund, presenting nearly 100 innovative products from over 30 cultural technology companies [3][4] Group 2 - The exhibition includes a digital art showcase of traditional Chinese auspicious patterns, featuring multimedia expressions and DIY workshops [3] - Advanced robotics and intelligent systems are prominently displayed, including soft robotic fish and autonomous flying robots [4] - The China Academy of Art presents an AIGC art domain, focusing on the creative applications of AI in various fields such as film, education, and gaming [4]

9月22日至25日，由联合国教科文组织主办，中国科学院与浙江省人民政府联合承办的第五届世界生物 圈保护区大会在杭州举行，来自全球150多个国家和地区的近4000名代表参会。 这是世界生物圈保护区大会首次在亚洲举办，本届大会以"塑造人与自然可持续的未来"为主题，展示我 国科技支撑生态文明建设的重要成就，为全球环境治理和推动可持续发展注入新动能。 "中国为'人与生物圈计划'作出了重要贡献，经验和做法值得推广。"联合国教科文组织"人与生物圈计 划"秘书长安东尼奥·艾伯鲁表示，中国展现了卓越的领导力，建立了全球规模最大的生物圈保护区国家 网络，其中的许多保护区已成为平衡环境保护与社会发展的全球典范。 中国科学院作为中国人与生物圈国家委员会主席单位，为生物多样性保护与可持续发展提供了强有力的 科技支撑。中国科学院院长侯建国表示，中国将继续与国际社会携手并肩，为实现联合国可持续发展目 标和共同构建地球生命共同体而努力奋斗。 科技赋能生态环境治理 中国生物圈保护工作一直重视新技术的应用。"中国开发了集图像、声纹及视频于一体的动植物物种智 能识别技术，融合卫星遥感、无人机和移动智能终端，构建了'天空地一体化'智慧监测体系。" ...

Core Insights - Bionics is driving innovation in intelligent robotic systems, particularly through the design and development of biomimetic robotic fish inspired by the superior maneuverability and perception of fish in complex underwater environments [1][4] - Traditional visual and sonar sensing technologies face performance limitations in challenging underwater conditions, while fish possess a unique lateral line system that provides a new paradigm for underwater robotic perception [1][4] - The research team from Peking University has proposed a novel data-driven framework that integrates modal decomposition and physical modeling for self-motion state estimation in biomimetic robotic fish [1][4] Summary by Sections Introduction to Bionics and Robotic Fish - Bionics connects nature and engineering, leading to advancements in intelligent robotic systems [1] - Fish exhibit exceptional maneuverability and perception in complex underwater environments, inspiring the design of biomimetic robotic fish [1] Challenges in Underwater Robotics - Traditional sensing technologies struggle in low-light and acoustically noisy underwater environments [1] - The unique lateral line system of fish, which senses water flow and pressure changes, offers a new approach for underwater robots [1] Development of Artificial Lateral Line System (ALLS) - Researchers have developed ALLS to simulate fish perception capabilities, applicable in flow field sensing, obstacle avoidance, and multi-fish coordination [1][4] - Accurate estimation of self-motion states, such as speed and trajectory, is crucial for autonomous intelligent robots [1] Proposed Framework for Self-Motion State Estimation - The new framework combines Proper Orthogonal Decomposition (POD) with physical modeling to estimate self-motion states [2][6] - The method extracts dominant modes related to fish movement from spatiotemporal pressure data collected by artificial lateral line sensors [2][6] Results and Validation - The model demonstrates robustness and generalization across various oscillation parameters, fish morphologies, and complex flow fields [4][14] - The research provides an efficient and reliable autonomous perception strategy for underwater biomimetic robots [4][14] Optimization of Sensor Layout - The study reveals redundancy in sensor arrangements and proposes optimized sensor layout strategies to reduce system complexity [11][13] - The optimal sensor distribution is validated through experiments, enhancing the understanding of fluid dynamics principles [13] Generalization to Complex Scenarios - The proposed method shows strong generalization capabilities in dynamic conditions and across different fish models [14] - It maintains robust state perception even in complex flow fields with wake interference, showcasing practical value [14] Conclusion and Future Directions - This research not only offers a solution for motion state estimation in robotic fish but also provides insights into the sensory mechanisms of real fish [16] - The integration of natural intelligence with engineering practices paves the way for the future development of intelligent, autonomous, and collaborative underwater robots [16]