传统的无人系统人工智能技术重点研究人脑思维、感知和肌电反应等领域的科学发现和技术实现。免疫反应是生物体在面临病毒、细菌和天敌时保持生存和健康的独特生理机制，是智能系统技术研究领域的崭新视点。受动物应对病毒侵袭、环境剧变、天敌威胁等不利态势的免疫反应、保护自我和进化机制启发，提出无人系统在包含攻击、干扰、拒止、封锁、损伤、故障和博弈等恶劣环境和对抗模式下的生存安全问题，建立无人系统免疫智能技术的一般框架，架设无人系统和生物体之间免疫机制的桥梁。主要内容包括无人系统免疫智能技术的基本概念、反应机理、关键技术和研究框架，并分别从感知与诊断、适应与激励、学习与进化等技术层面进行了问题描述。最后，对免疫智能技术的未来研究方向和应用领域进行了展望。
The traditional artificial intelligence of the unmanned system puts research focus on scientific discovery and technical implementation in the fields of human thought, perception and electromyographic response. The immune reaction is a unique physiological mechanism such that the organism could retain survival and health facing with the virus, germ and natural enemy. The so-called immune reaction inspires the unmanned system community, which includes the reaction including coping with the virus invading, the abrupt change of the environment, and the threatening of the natural enemy. In this paper, the survival problems of the unmanned system in the face of the attack, disturbance, jamming, blockade, damage, failure, and gaming are investigated. An universal framework of the immune intelligence of the unmanned system is established. In addition, the gap between the organism and the unmanned system is bridged. Considering the basic concept, reaction mechanism, key techniques and general framework, our paper puts research attention on the unmanned system. The problem is addressed from the perspectives of the perception and diagnosis, adaption and inspiration as well as learning and evolution. Finally, several possible future research directions of the immune intelligence techniques are provided.
[1] YANG G Z, BELLINGHAM J, DUPONT P E, et al. The grand challenges of science robotics[J]. Science Robotics, 2018, 3:eaar7650.
[2] MAZZOLAI B, LASCHI C. A vision for future bioinspired and biohybrid robots[J]. Science Robotics, 2020, 5:eaba6893.
[3] TRIMMER B A. Metal or muscle? The future of biologically inspired robots[J]. Science Robotics, 2020, 5:eaba6149.
[4] MARIA A, ALI R. Towards the development of robot immune system:A combined approach involving innate immune cells and T-lymphocytes[J]. Biosystems, 2018, 172, 52-67.
[5] HU Y, LUO W. Robotic immune systems:An architecture[C]//IEEE Symposium Series on Computational Intelligence (SSCI). Piscataway:IEEE Press, 2018:1098-1103.
[6] 杨照华,祁振强,房建成.空中目标的免疫识别算法[J].光学精密工程,2006,14(5):922-928. YAGN Z Y,QI Z Q,FANG J C.Immune recognition algorithm and its application to air target detection and recognition[J].Optics and Precision Engineering,2006,14(5):922-928(in Chinese).
[7] GUO L, CAO S. Anti-disturbance control for systems with multiple disturbances[M]. Boca Raton:CRC Press, 2013. [194. GUO L,YU Z,ZHANG X,et al.Safety control system technologies for UAVs:Review and prospect[J].Science in China (Information Sciences), 2020, 50(2):184-194(in Chinese).
[9] YU X, ZHANG Y M. Sense and avoid technologies with applications to unmanned aircraft systems:Review and prospects[J]. Progress in Aerospace Sciences, 2015, 74:152-166.
[10] YUAN Y, YANG H, GUO L, et al. Attack analysis and design of networked control systems under attack[M].Boca Raton:CRC Press, 2019.
[11] BAYAR N, DARMOUL S, HAJRI-GABOUJ S, et al. Fault detection, diagnosis and recovery using artificial immune systems:A review[J]. Engineering Applications of Artificial Intelligence, 2015, 46:43-57.
[12] CASTRO L N D, ZUBEN F J V. An evolutionary immune network for data clustering[C]//Brazilian Symposium on Neural Networks. Piscataway:IEEE Press, 2000.
[13] DASGUPTA D, YU S, NINO F. Recent advances in artificial immune systems:Models and applications[J]. Applied Soft Computing, 2011, 11(2):1574-1587.
[14] CANHAM R, JACKSON A H, TYRRELL A. Robot error detection using an artificial immune system[C]//Conference on Evolvable Hardware. Piscataway:IEEE Press, 2003:199-207.
[15] 焦李成,杜海峰,刘芳,等.免疫优化计算学习与识别[M].北京:科学出版社,2006. JIAO L C,DU H F,LIU F,et al.Immune computation for optimization learning and recognition[J].Beijing:Science Press,2006(in Chinese).
[16] MASUTTI T A S, CASTRO L N D. A self-organizing neural network using ideas from the immune system to solve the traveling salesman problem[J]. Information Sciences, 2009, 179(10):1454-1468
CJA