随着无人机技术的逐渐成熟和传统救援局限性的突出，无人机技术已广泛应用于空中搜救、自然灾害救援、城市消防救援等应急救援领域。借鉴无人机成熟的传感监测系统以及其极高的环境适用度，可对井下救援关键特征参数以及地形特征进行快速收集侦测，侦测无人机可代替传统救援方式深入受灾区域获取信息并提供可能。为了保障侦测无人机顺利到达目标区域执行救援工作以及安全返航，对人工水母算法（JS）进行优化，解决了其在路径规划过程中容易陷入局部最优等问题，主要取得了以下成果：

为解决无人机搜索效率低下和路径规划不理想的问题，将JS算法与混沌映射（Logistic）、高斯变异、飞行扰动等方法相结合，结果表明无人机在静态境中能规划出安全且无碰撞的路径。通过模拟实验验证了改进算法的效果，在障碍物占比14.56%实验中，优化后人工水母算法（IJS）与遗传算法（GA）相比，路径长度、规划时间和拐点数量分别减少了2.67%、72.23%、47.4%；与粒子群算法（PSO）算法比，路径长度、规划时间和拐点数量分别减少了3.95%、66.11%、50%；与JS算法相比，路径长度、规划时间和拐点数量分别减少了1.36%、70.87%、28.57%；在障碍物占比32.2%实验中，IJS算法与GA算法比，路径长度、规划时间和拐点数量分别减少了2.67%、62.5%、15.4%；与PSO算法比，路径长度、规划时间和拐点数量分别减少了3.95%、55.6%、18.5%；与JS算法比，路径长度、规划时间和拐点数量分别减少了1.36%、57.3%、8.3%，表明IJS算法相较于传统算法有显著提升。

为了验证改进融合算法有效性，搭建无人机平台，并搭载鱼眼相机作为传感器，以室内场景作为实验场地，设置起点和终点，在飞行过程中设置路径点引导无人机前往终点，通过最终路径验证其算法有效性。实验结果显示，无人机能够根据实时采集的环境信息，动态调整飞行路径，顺利绕过障碍物，最终平稳地到达指定的终点。实验结果表明：改进后的人工水母算法在实际应用中具有一定的可行性和效率，但在处理极端环境条件下的性能仍有待进一步验证。

通过JS算法与APF人工势场法进行结合，构建井下模拟三维环境，对算法进行仿真模拟，设计针对危险区域路径规划评价函数，对路径进行评价。通过仿真实验可得，在静态障碍物模拟实验中，IJS评价值为0.8540，低于PSO 0.9450和JS 0.8918；在动态障碍物模拟实验中，IJS评价值为0.8767，低于PSO 0.9761和JS 0.9014。改进后的算法在路径安全性上优于其他算法，评价函数低于其他几种算法，证明其路径的安全性更高。构建四旋翼无人机平台，并搭载D435i双目深度相机，在井下环境中对改进的融合算法进行了实验验证。结果表明，融合算法展现了高效的搜索性能和平滑的路径特征，达到了预期的目标。

With the maturation of unmanned aerial vehicle (UAV) technology and the prominent limitations of traditional rescue methods, UAV technology has been widely applied in fields such as aerial search and rescue, natural disaster response, and urban firefighting. Leveraging advanced sensor monitoring systems and high environmental adaptability, drones can rapidly collect and detect key rescue parameters and terrain features underground. Detection drones can replace traditional methods to penetrate disaster areas for information gathering and assistance. To ensure that detection drones successfully reach their target areas and return safely, the Jellyfish Search (JS) algorithm was optimized to address issues such as local optima in path planning, achieving significant improvements:

To address inefficiencies in UAV search and suboptimal path planning, the JS algorithm was combined with chaotic mapping (Logistic), Gaussian mutation, and flight perturbation methods. The results showed that UAVs could plan safe, collision-free paths in static environments. The effectiveness of the improved algorithm was verified through simulation experiments. In scenarios with 14.56% obstacle density, the optimized Jellyfish Search algorithm (IJS) reduced path length, planning time, and the number of turning points by 2.67%, 72.23%, and 47.4% respectively compared to the Genetic Algorithm (GA); by 3.95%, 66.11%, and 50% respectively compared to the Particle Swarm Optimization (PSO) algorithm; and by 1.36%, 70.87%, and 28.57% respectively compared to the original JS algorithm. In scenarios with 32.2% obstacle density, IJS showed reductions of 2.67% in path length, 62.5% in planning time, and 15.4% in turning points compared to GA; 3.95%, 55.6%, and 18.5% respectively compared to PSO; and 1.36%, 57.3%, and 8.3% respectively compared to JS, indicating significant improvements over traditional algorithms.

To verify the effectiveness of the improved hybrid algorithm, a UAV platform was equipped with a fisheye camera as the sensor, and indoor scenes were used as experimental sites. Starting and ending points were set, and path points were established during flight to guide the UAV to the destination. The final path validated the algorithm's effectiveness, showing that the UAV could dynamically adjust its flight path based on real-time environmental data, successfully navigating around obstacles to reach the designated endpoint smoothly. The results demonstrated that the improved Jellyfish Search algorithm has practical feasibility and efficiency in real-world applications, although its performance in extreme environmental conditions still requires further validation.

By integrating the JS algorithm with the Artificial Potential Field (APF) method, a simulated three-dimensional underground environment was created to evaluate the path planning through a designated evaluation function. In static obstacle simulations, the IJS scored 0.8540, lower than PSO's 0.9450 and JS's 0.8918; in dynamic obstacle simulations, IJS scored 0.8767, lower than PSO's 0.9761 and JS's 0.9014. The improved algorithm surpassed other methods in terms of path safety, as indicated by lower evaluation scores, proving its higher safety in path planning. A quadrotor UAV platform equipped with a D435i stereo depth camera was used in underground environments to experimentally validate the improved hybrid algorithm. The results showed that the hybrid algorithm exhibited efficient search performance and smooth path characteristics, achieving the anticipated goals.

[1]夏双双.煤矿井下微小型无人机路径规划和定位算法研究[D].青岛:山东科技大学,2018.

[2]Hart P.E, Nilsson N.J, Bertram R. Correction to "A Formal Basis for the Heuristic Determination of Minimum Cost Paths"[J]. ACM SIGART Bulletin, 1972(37): 420-426.

[3]Thi T, Cosmin C, Duc T, et al. A hierarchical global path planning approach for mobile robots based on multi-objective particle swarm optimization[J]. Applied Soft Computing, 2017(59): 15-26.

[4]XIE Z, ZHANG Q, JIANG Z, et al. Robot learning from demonstration for path planning: A review[J]. Science China(Technological Sciences), 2020, 63(08): 1325-1334.

[5]景裕,曹育森,朱明明,等.非接触式生命探测技术研究现状与发展[J].中国医疗设备,2021,36（6）:1-4.

[6]马骁.面向灾后救援的人体生命体征探测研究[D].成都:电子科技大学,2016.

[7]梁执桓,金婉妍.基于新型谱减算法的音频微振生命探测仪系统设计[J].电子制作,2018（增刊2）:16-18.

[8]陈湘源.煤矿无线通信系统的现状与发展[J].煤炭工程, 2009 (F01): 62-65.

[9]Jimoh A Y, Ojo O J. Rock-Eval pyrolysis and organic petrographic analysis of the Maastrichtian coals and shales at Gombe, Gongola Basin, Northeastern Nigeria[J]. Arabian Journal of Geosciences, 2016, 9(6): 1-13

[10]Raval S, Banerjee B P, Singh S K, et al. A preliminary investigation of mobile mapping technology for underground mining[C]//IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2019: 6071-6074.

[11]白琪. 基于机器视觉的矿井救灾机器人路径规划的研究[D].北京:中国矿业大 学,2018.

[12]孙海波.基于机器视觉的矿井环境三维地图构建研究[D].北京:中国矿业大学,2020.

[13]徐靖雯.基于机器视觉的井壁扫描成像系统研究[D].徐州:中国矿业大学,2021.

[14]郑学召,李诚康,文虎,等.矿井灾害救援生命信息探测技术及装备综述[J].煤矿安全,2017,48(12):116-119.

[15]王宁.基于FMCW雷达的多人定位及生命体征探测技术研究[D].南京:南京理工大学,2021.

[16]郑学召,孙梓峪,郭军,等.矿山钻孔救援多源信息探测技术研究与应用[J].煤田地 质与勘探,2022,50（11）:94-102.

[17]李钰梁.矿井超宽带雷达生命探测系统的研究[D].阜新:辽宁工程技术大学,2022.

[18]申祖辉.基于STM32的四旋翼无人机控制系统设计[D].洛阳:河南科技大学, 2019.

[19]王世勇.六旋翼无人机控制系统设计与实现[D].南京:南京航天航空大学, 2015.

[20]岳基隆,张庆杰, 朱华勇. 微小型四旋翼无人机研究进展及关键技术浅析[J]. 电光与控制, 2010, 17(010): 46-52.

[21]于力,高明昊,龚杰.国外救援无人机应用需求及发展趋势分析[J].飞航导弹,2018(04):33-36+82.

[22]Zhong Q, Fang X, Sharma V. A BigBiGAN-Based Loop Closure Detection Algorithm for Indoor Visual SLAM[J]. Journal of Electrical and Computer Engineering, 2021, 2021(06): 12-19.

[23]Min Haitao. Autonomous driving path planning algorithm based on improved A* algorithm in unstructured environment[J]. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 2021, 235(2): 129-145.

[24]Silva C,Tonidandel F. DVG+A* and RRT Path-Planners: A Comparison in a Highly Dynamic Environment[J]. Journal of Intelligent & Robotic Systems, 2021, 101(3): 33-39.

[25]KHATIB O.Real-time obstacle avoidance for manipulators and mobile robots[C]//IEEE International Conference on Robotics and Automation1985:500-505.

[26]Xian Y. Improvement and Fusion of A* Algorithm and Dynamic Window Approach

[27]Considering Complex Environmental Information[J]. Arabian Journal for Science and Engineering, 2021, 34: 139-150.

[28]Liang X. Double ant colony algorithm based on dynamic feedback for energy-saving route planning for ships[J]. Soft Computing, 2021, 25: 12-127.

[29]STENTZ A.The focused D*algorithm for real time replanning[C]//Proceedings of the 14th International Joint Conference on Artificial Intelligence,1995:1652-1659.

[30]STENTZ A.Optimal and efficient path planning for partially-known environments[C]//Proceedings of the 1994 IEEE International Conference onRobotics and Automation ,1994:3310-3317.

[31]KOENIG S,LIKHACHEV M.D*lite[C]// Proceedings of AAAI2002:476-483.

[32]陈侠,刘冬.应用D*Lite算法的目标移动时无人机三维路径规划[J].电光与控制201320(7):1-5.

[33]KOENIG S,LIKHACHEV M,FURCY D.Lifelong planning A*[J].Artificial Intelligence2004,155(1/2):93-146.

[34]槐创锋,郭龙,贾雪艳,等.改进A*算法与动态窗口法的机器人动态路径规划[J].计算机工程与应用,2021,57(08): 244-248.

[35]孔继利,张鹏坤,刘晓平.双向搜索机制的改进A*算法研究[J].计算机工程与应用,2021,57(08):231-237.

[36]陈娇,徐菱,陈佳,等.改进A*和动态窗口法的移动机器人路径规划[J].计算机集成制造系统,2021,55(06) :137-152.

[37]王中玉,曾国辉,黄勃,等.改进A~*算法的机器人全局最优路径规划[J].计算机应用,2019,39(09):2517-2522.

[38]张志文,张鹏,毛虎平,等.改进A*算法的机器人路径规划研究[J].电光与控制,2021,28(04):21-25.

[39]王维,裴东,冯璋.改进A~*算法的移动机器人最短路径规划[J].计算机应用,2018,38(05):1523-1526.

[40]LAVALLE S M,KUFFNERJ J.Randomized kinodynamic planning[C]//IEEE International Conference on Robotics and Automation,1999:473-479.

[41]KARAMAN S,FRAZZOLI E.Sampling-based algorithms for optimal motion planning[J].International Journal of RoboticsResearch,2011,30(7):846-894.

[42]AGUILARW G,MORALES S,RUIZ H,et al.RRT*GL based optimal path planning for real-time navigation of UAV[C]//International Work-Conference on Artificial Neural Networks2017:585-595.

[43]LIN Y C,SARIPALLI S.Sampling-based path planning for UAV collision avoidance[J].IEEE Transactions on Intelligent Transportation Systems,2017,18(11):3179-3192.

[44]MECHALI O,XU L M,WEI M Z,et al.A rectified RRT*with efficient obstacles avoidance method for UAV in 3Denvironment[C]//IEEE 9th Annual International Conference on CYBER Technology in Automation Control and Intelligent Systems,2019:480-485.

[45]张宇迪,龚鹏,胡为.改进人工势场法的智能车路径规划[J/OL].机械科学与技术:1-7[2023-03-06].

[46]LIU J Y,GUO Z Q,LIU S Y.The simulation of the UAV collision avoidance based on the artificial potential field method[J].Advanced Materials Research,2012,591/592/593:1400-1404.

[47]BUDIYANTO A,CAHYADI A,ADJI T B,et al.UAV obstacle avoidance using potential field under dynamic environment.[C]//International Conference on Control ElectronicsRenewable Energy and Communications(ICCEREC),2015:187-192.

[48]熊超,解武杰,董文瀚.基于碰撞锥改进人工势场的无人机避障路径规划[J].计算机工程,2018,44(9):314-320.

[49]张松灿,普杰信,司彦娜,等.蚁群算法在移动机器人路径规划中的应用综述[J]．计算机工程与应用,2020,56(08):10-19.

[50]胡佳斌,王祥澍,张琪,等.优化多步长蚁群算法求解机器人路径规划问题[J].传感器与微系统, 2021, 40(10): 121-124.

[51]李理,李鸿,单宁波.多启发因素改进蚁群算法的路径规划[J].计算机工程与应用,2019,55(05):219-225+250.

[52]毛嘉琪.改进蚁群算法的机器人路径规划研究[J].计算机应用与软件,2021,38(05):300-306.

[53]YANG L, FU L, LI P, et al. LF-ACO: an effective for mation path planning for multi-mobile robot[J]. Mathematical Biosciences and Engineering, 2022, 19(1): 225-252.

[54]袁福龙,朱建平.基于改进蚁群算法的移动机器人最优路径规划[J].现代制造工程,2021(07):38-47+65.

[55]LUO Q, WANG H, ZHENG Y, et al. Research on path planning of mobile robot based on improved ant colony algorithm[J]. Neural Computing and Applications, 2020, 32(6): 1555-1566.

[56]WANG P L, ZHANG T, XIAO Y J. Application research of ant colony cellular optimization algorithm in popula- tion evacuation path planning[J]. Acta Physica Sinica, 2020, 69(8).

[57]SUN S L, ZHAO Q, XIE W Z. Study on safe evacuation routes based on crowd density map of shopping mall[J]. IEEE Access, 2020, 8: 153981-153992.

[58]曹祥红,李欣妍,魏晓鸽,等.基于Dijkstra-ACO混合算法的应急疏散路径动态规划[J].电子与信息学报, 2020, 42(06): 1502-1509.

[59]李书霞. 改进群智能算法及其在PID控制器参数整定中 的应用[D].辽宁:辽宁科技大学,2016.

[60]王少泉,孙连英,张德政.基于创新理论的矿山灾害应对决策技术研究[J].金属矿山,2006.12:57-66.

[61]Sadhu A, Mukhopadhyaya S. Assembling of Robots in Presence of Line Obstacles with Direction-only Axes Agreement[J]. Journal of Physics: Conference Series, 2021, 1950(1): 18-25.

[62]闫国杰.矿山地质灾害研究与防治探讨[J].中国矿业,2004,13(3):66-68.

[63]蔺志伟,李奇敏,汪显宇.基于改进Census变换的单目视觉里程计[J].浙江大学学报(工学版),2021,55(8):1500-1509.

[64]张涛,张晨,魏宏宇.动态环境下融合激光雷达和IMU的激光里程计设计[J].导航定位与授时,2022,9(3):70-78.

[65]许智宾,李宏伟,张斌,等.双目视觉与惯导融合的移动机器人定位方法[J].测绘学报,2021,50(11):1512-1521.

[66]BEHLEY J, STACHNISS C. Efficient Surfel-Based SLAM using 3D Laser Range Data in Urban Environments[C]//Robotics: Science and Systems. 2018, 2013: 59.

[67]葛世荣,胡而已,裴文良.煤矿机器人体系及关键技术[J].煤炭学报,2020,45(1):455-463.

[68]Siddiqui Nimra Idris et al. Artificial Jellyfish Search Algorithm-Based Selective Harmonic Elimination in a Cascaded H-Bridge Multilevel Inverter[J]. Electronics, 2021, 10(19): 2402-2402

[69]Zhaohua L, Liu Z, Cui Z, et al. Parameter Optimization Analysis of Gmapping Algorithm Based on Improved RBPF Particle Filter[J]. Journal of Physics: Conference Series, 20 20, 1646(1): 598-611.

[70]Philo John Mark. Tacitus, Hector Boece, and the Writing of Scottish History[J]. Scottish Literary Review, 2020, 12(2): 122-127.

[71]Anonymous. SRI International; SRI International Releases New Version of Karto Software for Mobile Robot Navigation[J]. Robotics & Machine Learning, 2009, 9(9): 598-611.

[72]刘雪生.波分复用全固态激光雷达三维成像技术研究[D].保定:河北大学,2021:43-50.

[73]许雪梅.基于模拟退火算法改进遗传算法的织物智能配色[J].纺织学报,2021,42(07):123-128.

[74]李俊涛,毛红保,张鹏,等.基于多优化策略RRT的无人机实时航线规划[J].火力与指挥控制,2017,42(12):115-119+124.

[75]徐力,刘云华,王启富.自适应遗传算法在机器人路径规划的应用[J].计算机工程与应用,2020,56(18):36-41.

[76]蔡开泉,唐焱武,张学军,等.An improved genetic algorithm with dynamic topology[J]. Chinese Physics B,2016,25(12):587-593.

[77]付留芳,周明,李文哲,等.基于遗传算法的UUV应召搜潜路径规划[J/OL].电光与控制,1-9[2024-04-21].

[78]郭利进,李强.基于改进RRT*算法的移动机器人路径规划[J/OL].智能系统学报,1-8[2024-03-23].

[79]谢春圆,王欣,吴迪,等.复杂环境下的改进RRT算法路径规划[J].国外电子测量技术,2024,43(02):131-138.

[80]彭君.改进RRT算法在移动机器人路径规划中的应用研究[D].南京:南京邮电大学,2022.

[81]刘成菊,韩俊强,安康.基于改进RRT算法的RoboCup机器人动态路径规划[J].机器人,2017,39(01):8-15.

[82]王全.基于RRT的全局路径规划方法及其应用研究[D].长沙:国防科学技术大学,2014.

[83]宋晓琳,周南,黄正瑜等.改进RRT在汽车避障局部路径规划中的应用[J].湖南:湖南大学学报(自然科学版),2017,44(04):30-37.

[84]Bizhao P, Wei D, Xin H, Fu D, et al. Multiple air route crossing waypoints optimization via artificial potential field method[J]. Chinese Journal of Aeronautics, 2021, 34(04):27 9-292.

[85]潘红丽.基于改进粒子群算法的垃圾清运车辆低碳路径规划[D].南京:南京信息工程大学,2022.

[86]张津源.基于改进粒子群算法的物流路径规划研究[D].哈尔滨:哈尔滨师范大学,2021.

[87]曾胜昆.融合激光雷达与视觉的环境感知和局部路径规划算法研究[D].赣州市:赣南师范大学,2023.

[88]范佳哲.基于环境信息的局部路径规划算法改进与应用[D].杭州电子科技大学,2023.

[89]于军琪,陈易圣,冯春勇,等.智能建造机器人局部路径规划研究综述[J/OL].计算机工程与应用,1-16[2024-04-21].

[90]原新,苑新伟,刘志林.基于蝙蝠算法和动态窗口法的混合路径规划[J].实验技术与管理,2021,38(10):30-36.

[91]张国胜,李彩虹,张耀玉,等.基于改进人工势场法的机器人局部路径规划[J/OL].计算机工程,1-9[2024-04-21].

[92]梅艺林,崔立堃,胡雪岩.基于人工势场法的无人车路径规划与避障研究[J/OL].兵器装备工程学报,1-6[2024-04-21].

[93]周世博.基于改进的A*算法和人工势场法的轮式机器人路径规划研究[D].太原:太原理工大学,2021.

[94]Jui-Sheng Chou et al., A novel metaheuristic optimizer inspired by behavior of jellyfish in ocean. Applied Mathematics and Computation.

[95]王向宇.基于改进A~*算法和动态窗口法的室内移动机器人路径规划研究[D].赣州:江西理工大学,2023.

[96]迟旭.基于改进A*算法与动态窗口法融合的移动机器人路径规划研究[D].大连:大连交通大学,2021.

[97]王彬,聂建军,李海洋,等.优化A*与动态窗口法的移动机器人路径规划[J/OL].计算机集成制造系统,1-17[2024-04-21].

[98]贺勇,侯体成,曾子望.融合改进A*和动态窗口法的无人机路径规划[J/OL].机械科学与技术,1-9[2024-04-21].

[99]褚宏悦,易军凯.无人机安全路径规划的混沌粒子群优化研究[J/OL].控制工程,1-8[2024-03-23].

[100]肖义平,赵云峰.基于混沌映射和高斯扰动的改进粒子群算法MPPT控制策略研究[J/OL].电源学报,1-13[2024-03-23].

[101]邵长春,胡国明,李水明.焊接机器人路径规划的高斯变异蝗虫优化算法[J].机械设计与制造,2023(04):276-280.

[102]李永迪,李彩虹,张耀玉等.基于APF-LSTM-DDPG算法的移动机器人局部路径规划[J].山东理工大学学报(自然科学版),2024,38(01):33-41.

[103]林俊志,席万强,周俊等.基于改进PRM和APF的移动机器人路径规划[J].国外电子测量技术,2022,41(12):1-6.

[104]陈丹,姚伯羽,吴欣.基于ROS的移动机器人核相关目标跟踪方法的优化设计[J].计算机辅助设计与图形学学报,2020,32(12):1967-1975.

[105]毛新军.自主机器人软件工程的研究综述[J].计算机学报,2021,44(08):1661-1678.

[106]冯德胜.基于ROS的移动机器人导航系统研究[D].济南:齐鲁工业大学,2021:13-26.

[107]任超锋,张楠.数码相机畸变模型的相互转换方法[J].激光与光电子学进展,2018, 55(7):406-413.

[108]Zhang Z. Flexible camera calibration by viewing a plane from unknown orientations [C]//Proceedings of the seventh ieee international conference on computer vision. IEEE,1999, 1: 666-673.

[109]Furgale P, Rehder J, Siegwart R. Unified temporal and spatial calibration for multi-sensorsystems[C]//2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.IEEE, 2013: 1280-1286.

[110]Qin T, Shen S. Online temporal calibration for monocular visual-inertial systems[C]//2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2018:3662-3669.