Large-Scale Swarm Control in Cluttered Environments
المؤلف | Elsayed, Saber |
المؤلف | Mabrok, Mohamed |
تاريخ الإتاحة | 2024-02-28T05:17:08Z |
تاريخ النشر | 2024-01-01 |
اسم المنشور | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
المعرّف | http://dx.doi.org/10.1007/978-981-99-8715-3_32 |
الاقتباس | Elsayed, S., & Mabrok, M. (2023, December). Large-Scale Swarm Control in Cluttered Environments. In International Conference on Social Robotics (pp. 384-395). Singapore: Springer Nature Singapore. |
الترقيم الدولي الموحد للكتاب | 9789819987146 |
الرقم المعياري الدولي للكتاب | 03029743 |
الملخص | In the evolving era of social robots, managing a swarm of autonomous agents to perform particular tasks has become essential for numerous industries. The task becomes more challenging for large-scale swarms and complex environments, which have not been fully explored yet. Therefore, this research introduces a methodology incorporating multiple coordinated robotic shepherds to effectively guide large-scale agent swarms in obstacle-laden terrains. The proposed framework commences with deploying an unsupervised machine-learning algorithm to categorise the swarm into clusters. Then, a shepherding algorithm with coordinated robotic shepherds drives the sub-swarms towards the goal. Also, a path planner based on an evolutionary algorithm is proposed to help robotic shepherds move in a way that minimises the dispersion of each sub-swarm and avoids potential hazards and obstructions. The proposed approach is tested on different scenarios, with the results showing a success rate of 100% in guiding swarms with sizes up to 3000 agents. |
اللغة | en |
الناشر | Springer Science and Business Media Deutschland GmbH |
الموضوع | Large-scale Path Planning Robotic Shepherding Swarm Control |
النوع | Conference Paper |
رقم المجلد | 14453 LNAI |
الملفات في هذه التسجيلة
الملفات | الحجم | الصيغة | العرض |
---|---|---|---|
لا توجد ملفات لها صلة بهذه التسجيلة. |
هذه التسجيلة تظهر في المجموعات التالية
-
الرياضيات والإحصاء والفيزياء [726 items ]