专题:培育新质生产力 助力高水平科技自立自强

自主无人系统的具身认知智能框架

  • 孙长银 ,
  • 穆朝絮 ,
  • 柳文章 ,
  • 王晓
展开
  • 1. 安徽大学, 合肥 230039;
    2. 自主无人系统技术教育部工程研究中心, 合肥 230039;
    3. 安徽大学人工智能学院, 合肥 230039;
    4. 安徽省无人系统与智能技术工程研究中心, 合肥 230039
孙长银,教授,研究方向为自主无人系统控制,电子信箱:cysun@ahu.edu.cn;王晓(通信作者),教授,研究方向为自主无人系统的社会认知智能,电子信箱:xiao.wang@ahu.edu.cn

收稿日期: 2024-05-01

  修回日期: 2024-06-12

  网络出版日期: 2024-07-09

基金资助

国家自然科学基金创新研究群体项目(61921004)

Embodied cognitive intelligence framework of unmanned autonomous systems

  • SUN Changyin ,
  • MU Chaoxu ,
  • LIU Wenzhang ,
  • WANG Xiao
Expand
  • 1. Anhui University, Hefei 230039, China;
    2. Engineering Research Center of Autonomous Unmanned System Technology, Ministry of Education, Hefei 230039, China;
    3. School of Artificial Intelligence, Anhui University, Hefei 230039, China;
    4. Anhui Provincial Engineering Research Center for Unmanned System and Intelligent Technology, Hefei 230039, China

Received date: 2024-05-01

  Revised date: 2024-06-12

  Online published: 2024-07-09

摘要

自主无人系统是一类具有自主认知、运动规划、自主决策和推理能力的智能系统,其目标是在有限甚至没有人工参与的情况下完成复杂开放动态场景中的通用任务。针对自主无人系统在跨域协同任务上往往面临协同感知效率低、自组网通信可靠性差、资源调度流程慢、任务分配易冲突等一系列问题,探讨了融合大模型和生成式人工智能技术,构建了“大模型+自主无人系统+人工智能生成内容”为一体的自主无人系统“算-控-测”具身认知智能架构,以推动自主无人系统具身认知智能应用落地。

本文引用格式

孙长银 , 穆朝絮 , 柳文章 , 王晓 . 自主无人系统的具身认知智能框架[J]. 科技导报, 2024 , 42(12) : 157 -166 . DOI: 10.3981/j.issn.1000-7857.2024.06.00703

Abstract

Unmanned autonomous systems (UASs) are intelligent systems endowed with autonomous cognition, motion planning, autonomous decision making and reasoning capabilities. Their goals are designed to perform and complete common tasks in complex, open and dynamic scenarios with limited or even no human participation. In terms of the challenges UASs faced in cross-domain collaborative tasks, such as low efficiency of collaborative perception, poor reliability of Ad Hoc network communication, slow resource scheduling, and conflict-prone task allocation, this paper explored how to combine large models and generative artificial intelligence (GAI) technology to construct the“compute-control-test”embodied cognitive intelligence framework of UASs integrating“large model + autonomous unmanned systems + artificial intelligence generated content(AIGC)”. It will provide valuable reference for advancing the technological implementation and practical deployment of UASs with embodied cognitive intelligence.

参考文献

[1] Wiener N. Cybernetics, or control and communication in the animal and the machine (2nd ed.)[M]. Cambridge:MIT Press, 1961.
[2] Wang F Y. A big-data perspective on AI:Newton, Merton, and analytics intelligence[J]. IEEE Intelligent Systems, 2012, 27(5):2-4.
[3] Varela F J, Thompson E, Rosch E. The embodied mind:Cognitive science and human experience[M]. Cambridge:MIT Press, 1991.
[4] Gupta A, Savarese S, Ganguli S, et al. Embodied intelligence via learning and evolution[J]. Nature Communications, 2021, 12:5721.
[5] Mengaldo G, Renda F, Brunton S L, et al. A concise guide to modelling the physics of embodied intelligence in soft robotics[J]. Nature Reviews Physics, 2022, 4(9):595-610.
[6] Zhang C, Chen J X, Li J T, et al. Large language models for human-robot interaction:A review[J]. Biomimetic Intelligence and Robotics, 2023, 3(4):100131.
[7] Mahadevan K, Chien J, Brown N, et al. Generative expressive robot behaviors using large language models[J/OL].[2024-05-01]. https://arxiv.org/pdf/2401.14673v2.
[8] Cao X, Sun C Y, Wang X R. Threat assessment strategy of human-in-the-loop unmanned underwater vehicle under uncertain events[J]. IEEE Transactions on Systems, Man, and Cybernetics:Systems, 2024, 54(1):520-532.
[9] Wang Y D, Cao J Y, Sun J, et al. Path following control for unmanned surface vehicles:A reinforcement learningbased method with experimental validation[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024:1-14.
[10] Wang F Y, Zheng N N, Li L, et al. China's 12-year quest of autonomous vehicular intelligence:The intelligent vehicles future challenge program[J]. IEEE Intelligent Transportation Systems Magazine, 2021, 13(2):6-19.
[11] Bektaş K, Thrash T, van Raai M A, et al. The systematic evaluation of an embodied control interface for virtual reality[J]. PLoS One, 2021, 16(12):e0259977.
[12] Miao Q H, Lv Y S, Huang M, et al. Parallel learning:Overview and perspective for computational learning across Syn2Real and Sim2Real[J]. IEEE/CAA Journal of Automatica Sinica, 2023, 10(3):603-631.
[13] Dong L, He Z C, Song C W, et al. Multi-robot socialaware cooperative planning in pedestrian environments using attention-based actor-critic[J]. Artificial Intelligence Review, 2024, 57(4):108.
[14] Rosenblueth A, Wiener N, Bigelow J. Behavior, purpose and teleology[J]. Philosophy of Science, 1943, 10(1):18-24.
[15] Okita S Y. Social interactions and learning[M]//Encyclopedia of the Sciences of Learning. Boston. MA:Springer US, 2012:3104-3107.
[16] Kuipers B, Feigenbaum E A, Hart P E, et al. Shakey:From conception to history[J]. AI Magazine, 2017, 38(1):88-103.
[17] Huang W L, Wang C, Zhang R H, et al. VoxPoser:Composable 3D value maps for robotic manipulation with language models[J]. arXiv e-prints, 2023, doi:10.48550/arXiv.2307.05973.
[18] Bartolozzi C, Indiveri G, Donati E. Embodied neuromorphic intelligence[J]. Nature Communications, 2022, 13:1024.
[19] Matsuo Y, LeCun Y, Sahani M, et al. Deep learning, reinforcement learning, and world models[J]. Neural Networks, 2022, 152:267-275.
[20] Liang J, Huang W L, Xia F, et al. Code as policies:Language model programs for embodied control[C]//Proceedings of IEEE International Conference on Robotics and Automation (ICRA). Piscataway, NJ:IEEE, 2023:9493-9500.
[21] Wang J G, Wang X, Tian Y L, et al. Parallel training:An ACP-based training framework for iterative learning in uncertain driving spaces[J]. IEEE Transactions on Intelligent Vehicles, 2023, 8(4):2832-2841.
[22] Wang X, Yang J, Han J P, et al. Metaverses and DeMetaverses:From digital twins in CPS to parallel intelligence in CPSS[J]. IEEE Intelligent Systems, 2022, 37(4):97-102.
[23] Gan Y H, Zhang B, Shao J W, et al. Embodied intelligence:Bionic robot controller integrating environment perception, autonomous planning, and motion control[J]. IEEE Robotics and Automation Letters, 2024, 9(5):4559-4566.
[24] Fei N Y, Lu Z W, Gao Y Z, et al. Towards artificial general intelligence via a multimodal foundation model[J]. Nature Communications, 2022, 13:3094.
[25] Kalinowska A, Pilarski P M, Murphey T D. Embodied communication:How robots and people communicate through physical interaction[J]. Annual Review of Control, Robotics, and Autonomous Systems, 2023, 6:205-232.
[26] Xu L L, Wang T, Wang J W, et al. Attention-based policy distillation for UAV simultaneous target tracking and obstacle avoidance[J]. IEEE Transactions on Intelligent Vehicles, 2024, 9(2):3768-3781.
[27] Zhang B, Zhu J, Su H. Toward the third generation artificial intelligence[J]. Science China Information Sciences, 2023, 66(2):121101.
[28] 王飞跃, 缪青海. 基础智能:从联邦智能到基于TAO的智能系统联邦[J]. 科技导报, 2023, 41(19):103-112.
[29] 缪青海, 王兴霞, 杨静, 等. 从基础智能到通用智能:基于大模型的GenAI和AGI之现状与展望[J]. 自动化学报, 2024, 50(4):674-687.
[30] Mu Y, Yao S Y, Ding M Y, et al. EC2:Emergent communication for embodied control[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ:IEEE, 2023:6704-6714.
[31] Li L, Wang X, Wang K F, et al. Parallel testing of vehicle intelligence via virtual-real interaction[J]. Science Robotics, 2019, 4(28):eaaw4106.
[32] Sun C Y, Liu W Z, Dong L. Reinforcement learning with task decomposition for cooperative multiagent systems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(5):2054-2065.
[33] Liu W Z, Cai W Z, Jiang K, et al. XuanCe:A comprehensive and unified deep reinforcement learning library[J/OL].[2024-05-01]. https://arxiv.org/html/2312.16248v 1.
[34] Sun C, Wu X, Wang Y D, et al. Attention-based value classification reinforcement learning for collision-free robot navigation[J]. IEEE Transactions on Intelligent Vehicles, 2024, doi:10.1109/TIV.2024.3391007.
文章导航

/