在智能安防、搜索救援、环境监测等领域，与单体机器人系统相比，多机器人系统具有更强的容错能力、更好的适应性和资源的高效利用率，多机器人系统对系统体系结构设计、运动规划与协作编队控制算法均提出新要求。

多机器人系统采用分层式结构协同体系，对A*算法改进并与带有机器人及环境约束的TEB（Time Elastic Band）方法相结合使单机器人具有运动规划能力，采用交通规则法解决多机器人冲突问题，针对多机器人协同编队设计两轮差动转向多机器人编队控制算法。主要完成以下工作内容：

（1）针对多机器人系统体系与结构，以分层式结构解决传统结构资源过集中、通信复杂的问题；在机械结构上采用两轮差动转向移动平台；硬件结构上采用Kobuki+TX2混合结构进行层次化设计，在充分利用不同层级计算资源的同时增强机器人硬件可扩展性；软件结构上采用ROS机器人操作系统易于多机器人程序开发与部署。

（2）针对多机器人运动规划，传统A*路径规划算法拐点多、不平滑，缺乏动态避障，导致机器人行驶时间长，不适用多机器人运动等问题。在传统A*算法估价函数中引入拐角代价值和起点朝向代价值；对两轮差动转向机器人与环境进行分析得到路径点与障碍约束、速度与加速度约束、机器人运动学约束及时间最优约束，将改进A*算法所生成路径用于带有以上约束的TEB方法中进行实时动态路径规划，使机器人行驶时间短、路径平滑、具有动态避障能力且可直接输出控制参数便于机器人反馈控制；针对多机器人系统运动规划冲突问题，采用交通规则法充分利用各机器人自主决策能力解决机器人之间冲突问题。最后通过Kobuki平台与MATLAB仿真平台验证运动规划算法有效性。

（3）针对多机器人系统编队控制，设计一种基于两轮差动转向的多机器人编队控制算法。建立机器人运动学模型，针对经典三角形编队队形进行分析得到链式系统与误差模型，正则变换后根据李亚普洛夫函数反推出全局跟踪控制律以实现机器人编队精准控制。最后在MATLAB与Gazebo仿真平台引入本文所设计算法进行仿真，仿真结果表明所设计编队算法可以快速进行收敛至稳定状态。

In the fields of intelligent security, search and rescue, environmental monitoring, etc., compared with a single robot system, the multi-robot system has stronger fault tolerance, better adaptability and efficient utilization of resources. The multi-robot system is designed for the system architecture , Motion planning and cooperative formation control algorithms all put forward new requirements.

The multi-robot system adopts a hierarchical structure and coordination system, and proposes a path planning algorithm that combines an improved A* algorithm and a TEB (Time Elastic Band) method with robots and environmental constraints, so that a single robot has motion planning capabilities and adopts traffic rules. The method solves the robot conflict problem when multi-robots are moving, and designs a two-wheel differential steering multi-robot formation control algorithm for multi-robot cooperative formation. Mainly complete the following work content:

(1) Aiming at the multi-robot system system and structure, use a layered structure to solve the problems of over-concentration of traditional structural resources and complex communication; use a two-wheel differential steering mobile platform in the mechanical structure; use a Kobuki+TX2 hybrid structure in the hardware structure Carry out hierarchical design to enhance the scalability of robot hardware while making full use of different levels of computing resources; the ROS robot operating system is adopted in the software structure to facilitate the development and deployment of multi-robot programs.

(2) For multi-robot motion planning, the traditional A* path planning algorithm has many inflection points, is not smooth, and lacks dynamic obstacle avoidance, which causes the robot to travel for a long time and is not suitable for multi-robot motion. Introduce the corner value and the starting point value in the traditional A* algorithm evaluation function; analyze the two-wheel differential steering robot and the environment to obtain the path point and obstacle constraints, speed and acceleration constraints, robot kinematics constraints, and time optimal constraints , The path generated by the improved A* algorithm is used in the TEB method with the above constraints for real-time dynamic path planning, so that the robot has short driving time, smooth path, dynamic obstacle avoidance ability, and direct output control parameters for robot feedback control; Aiming at the conflict problem of multi-robot system motion planning, the traffic rule method is adopted to make full use of the autonomous decision-making ability of each robot to solve the conflict problem between robots. Finally, the validity of the motion planning algorithm is verified by the Kobuki platform and the MATLAB simulation platform.

(3) Aiming at the formation control of the multi-robot system, a multi-robot formation control algorithm based on two-wheel differential steering is designed. Establish the robot kinematics model, analyze the classic triangle formation formation to obtain the chain system and error model, after the canonical transformation, the global tracking control law is deduced according to the Lyapulov function to realize the precise control of the robot formation. Finally, the algorithm proposed in this paper is introduced into the MATLAB and Gazebo simulation platform for simulation. The simulation results show that the designed formation algorithm can quickly converge to a stable state, which verifies the effectiveness and practicability of the algorithm.

[1] Parker, Lynne E., Daniela Rus, and Gaurav S. Sukhatme. "Multiple mobile robot systems." Springer Handbook of Robotics. Springer, Cham, 2016. 1335-1384.

[2] 日本机器人学会, 许郁文. 机器人科技[M]. 人民邮电出版社, 2015.

[3] Delmerico J, Mintchev S, Giusti A, et al. The current state and future outlook of rescue robotics[J]. Journal of Field Robotics, 2019, 36(7): 1171-1191.

[4] Cardarelli E, Digani V, Sabattini L, et al. Cooperative cloud robotics architecture for the coordination of multi-AGV systems in industrial warehouses[J]. Mechatronics, 2017, 45: 1-13.

[5] Troccaz J, Dagnino G, Yang G Z. Frontiers of medical robotics: from concept to systems to clinical translation[J]. Annual review of biomedical engineering, 2019, 21: 193-218.

[6] Yang G Z, Bellingham J, Dupont P E, et al. The grand challenges of Science Robotics[J]. Science robotics, 2018, 3(14): eaar7650.

[7] Abad P, Franco M, Castillón R, et al. Integrating an autonomous robot on a dance and new technologies festival[C]. Iberian robotics conference. Springer, Cham, 2017: 75-87.

[8] Zhang C, Noguchi N. Development of a multi-robot tractor system for agriculture field work[J]. Computers and Electronics in Agriculture, 2017, 142: 79-90.

[9] 王莹.浅谈工业机器人在工业生产中的应用[J].中国新技术新产品,2015,9(17):19.

[10] Petersen K H, Napp N, Stuart-Smith R, et al. A review of collective robotic construction[J]. Science Robotics, 2019, 4(28).

[11] Singh A, Sharma H, Singh N, et al. A survey on cooperative mobile robotics[J]. International Journal of Applied Engineering Research, 2018, 13(7): 5160-5166.

[12] Hamann H. Swarm robotics: A formal approach[M]. Berlin: Springer, 2018.

[13] Zhongyuan G U O, Wenjing Y, Minglong L I, et al. ALLIANCE-ROS: A software framework on ROS for fault-tolerant and cooperative mobile robots[J]. Chinese Journal of Electronics, 2018, 27(3): 467-475.

[14] Kapetanakis S , Ku De Nko D . Reinforcement Learning of Coordination in Heterogeneous Cooperative Multi-agent Systems[M]. Springer Berlin Heidelberg, 2005.

[15] Gil S, Kumar S, Katabi D, et al. Adaptive communication in multi-robotsystems using directionality of signal strength[J]. The International Journal of Robotics Research, 2015, 34(7): 946-968.

[16] 黄琰, 李岩, 俞建成,等. AUV智能化现状与发展趋势[J]. 机器人, 2020, 42(2):215-231.

[17] Chen W , Zhou S , Pan Z , et al. Mapless Collaborative Navigation for a Multi-Robot System Based on the Deep Reinforcement Learning[J]. Applied Sciences, 2019, 9(20):4198.

[18] Gustiana M, Indrawaty Y, Febriandi A. Perancangan Mobile Manipulator Robot Secara Simulasi Menggunakan Microsoft Robotics Developer Studio[J]. MIND Journal, 2018, 3(1): 15-23.

[19] Liu Y, Hu B, Zhang S, et al. Multi-robot Merged Formation Algorithm Simulation in Webots[C]. Proceedings of the 2020 International Conference on Aviation Safety and Information Technology. 2020: 560-562.

[20] Uslu E, Cakmak F, Altuntaş N, et al. An architecture for multi-robot localization and mapping in the Gazebo/Robot Operating System simulation environment[J]. Simulation, 2017, 93(9): 771-780.

[21] 吕云峰. 障碍环境中多机器人路径规划及任务分配[D]. 湖南大学, 2017.

[22] 岳伟韬, 苏婧, 谷志珉等. 占据栅格地图的最佳栅格大小与地图精度[J]. 机器人, 2020, 42(2):199-206.

[23] 蒋林, 程文凯, 朱志超等. 一种STEDF的可视图环境建模方法[J]. 机械设计与制造, 2018, 000(005):253-255.

[24] 孙弋、张笑笑. 结合退火优化和遗传重采样的RBPF算法[J]. 西安科技大学学报，2020，40(02)：349-355.

[25] Dorigo M, Birattari M, Stutzle T. Ant colony optimization[J]. IEEE computational intelligence magazine, 2006, 1(4): 28-39.

[26] 陈志，江治杰，刘瑶. 基于改进蚁群算法的不同路段低碳物流路径优化研究[J]. 生态经济，2019，35(12)：53-59+66.

[27] 张晓莉，杨亚新，谢永成. 改进的蚁群算法在机器人路径规划上的应用[J]. 计算机工程与应用，2020，56(02)：29-34.

[28] Cecilia J M , Garcia J M , Ujaldon M , et al. Parallelization Strategies for Ant Colony Optimisation on GPUs[J]. 2011.

[29] 郝冬, 刘斌. 基于模糊逻辑行为融合路径规划方法[J]. 计算机工程与设计, 2009, 30(3):660-663.

[30] Long P, Fan T, Liao X, et al. Towards optimally decentralized multi-robot collision avoidance via deep reinforcement learning[C], 2018 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2018: 6252-6259.

[31] Dasari S, Ebert F, Tian S, et al. Robonet: Large-scale multi-robot learning[J]. arXiv preprint arXiv:1910.11215, 2019.

[32] Mehdi Ghatee, Ali Mohades. Motion Planning in Order to Optimize the Length and Clearance Applying a Hopfield Neural Network. Expert Systems with Applications, 2009, 36(3):4688-4695

[33] Elterman L. UV, visible, and IR attenuation for altitudes to 50 km, 1968[R]. AIR FORCE CAMBRIDGE RESEARCH LABS HANSCOM AFB MA, 1968.

[34] 程传奇, 郝向阳, 李建胜等. 融合改进A*算法和动态窗口法的全局动态路径规划[J]. 西安交通大学学报, 2017, 51(011):137-143.

[35] 陈骏岭、秦小麟、李星罗等. 基于人工势场法的多机器人协同避障[J]. 计算机科学, 2020, v.47(11):228-233.

[36] 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, 9: 314-320.

[37] [任彦, 赵海波. 改进人工势场法的机器人避障及路径规划[J]. 计算机仿真, 2020, v.37(02):365-369.

[38] 苏治宝, 陆际联. 基于行为法队形保持中的队形反馈[J]. 机床与液压, 2003(03):167-169.

[39] 王伟. 基于多行为融合的多移动机器人协同路径规划研究[D]. 2017.

[40] Sun J , Tang J , Lao S . Collision Avoidance for Cooperative UAVs With Optimized Artificial Potential Field Algorithm[J]. IEEE Access, 2017, 5:18382-18390.

[41] 高力, 陆丽萍, 褚端峰等. 基于图与势场法的多车道编队控制[J]. 自动化学报, 2020, 46(1).

[42] Low C B. Adaptable virtual structure formation tracking control design for nonholonomic tracked mobile robots, with experiments[C]. 2015 IEEE 18th International Conference on Intelligent Transportation Systems. IEEE, 2015: 1868-1875.

[43] Abbasi Y , Moosavian S , Novinzadeh A B . Formation control of aerial robots using virtual structure and new fuzzy-based self-tuning synchr·onization[J]. Transactions of the Institute of Measurement and Control, 2016, 39(12).

[44] 冯之文, 姚尧, 苗艳等. 基于时延补偿的AUV领航跟随编队控制[J]. 舰船电子对抗, 2020, 043(001):29-36.

[45] 易国, 毛建旭, 王耀南等.非完整移动机器人领航-跟随编队分布式控制[J].仪器仪表学报,2017,38(09):2266-2272.

[46] 师五喜, 王健. 多移动机器人的协同编队控制[J]. 天津工业大学学报, 2020, 039(001):63-68.

[47] Jawhar I, Mohamed N, Wu J, et al. Networking of multi-robot systems: Architectures and requirements[J]. Journal of Sensor and Actuator Networks, 2018, 7(4): 52.

[48] Serebrenny V, Shereuzhev M. Dependence of dynamics of multi-robot system on control architecture[M]. Robotics: Industry 4.0 issues & new intelligent control paradigms. Springer, Cham, 2020: 125-132.

[49] Nair M, Givigi S. The impact of communications considerations in multi-robot systems[C]. 2019 International Conference on Communications, Signal Processing, and their Applications (ICCSPA). IEEE, 2019: 1-6.

[50] Zieliński C, Winiarski T, Kornuta T. Agent-based structures of robot systems[C].Polish Control Conference. Springer, Cham, 2017: 493-502.

[51] Abdelkader M. Towards Real-Time Distributed Planning in Multi-Robot Systems[D]. , 2018.

[52] Mohamed K, El Shenawy A, Harb H. A hybrid decentralized coordinated approach for multi-robot exploration task[J]. The Computer Journal, 2019, 62(9): 1284-1300.

[53] Wan S, Lu J, Fan P. Semi-centralized control for multi robot formation[C]. 2017 2nd International Conference on Robotics and Automation Engineering (ICRAE). IEEE, 2017: 31-36.

[54] 王辉. 基于ROS的机器人路径导航系统的设计与实现[D]. 中国科学院大学.