Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordination
Abstract This work analyses the performance of Reinforcement Learning (RL) versus Swarm Intelligence (SI) for coordinating multiple unmanned High Altitude Platform Stations (HAPS) for communications area coverage. It builds upon previous work which looked at various elements of both algorithms. The...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2021-05-01
|
| Series: | SN Applied Sciences |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s42452-021-04658-6 |
| _version_ | 1824046158454980608 |
|---|---|
| author | Ogbonnaya Anicho Philip B. Charlesworth Gurvinder S. Baicher Atulya K. Nagar |
| author_facet | Ogbonnaya Anicho Philip B. Charlesworth Gurvinder S. Baicher Atulya K. Nagar |
| author_sort | Ogbonnaya Anicho |
| collection | DOAJ |
| description | Abstract This work analyses the performance of Reinforcement Learning (RL) versus Swarm Intelligence (SI) for coordinating multiple unmanned High Altitude Platform Stations (HAPS) for communications area coverage. It builds upon previous work which looked at various elements of both algorithms. The main aim of this paper is to address the continuous state-space challenge within this work by using partitioning to manage the high dimensionality problem. This enabled comparing the performance of the classical cases of both RL and SI establishing a baseline for future comparisons of improved versions. From previous work, SI was observed to perform better across various key performance indicators. However, after tuning parameters and empirically choosing suitable partitioning ratio for the RL state space, it was observed that the SI algorithm still maintained superior coordination capability by achieving higher mean overall user coverage (about 20% better than the RL algorithm), in addition to faster convergence rates. Though the RL technique showed better average peak user coverage, the unpredictable coverage dip was a key weakness, making SI a more suitable algorithm within the context of this work. |
| first_indexed | 2024-12-23T00:30:52Z |
| format | Article |
| id | doaj.art-0316747f277543a1b24c8ba84a999016 |
| institution | Directory Open Access Journal |
| issn | 2523-3963 2523-3971 |
| language | English |
| last_indexed | 2024-12-23T00:30:52Z |
| publishDate | 2021-05-01 |
| publisher | Springer |
| record_format | Article |
| series | SN Applied Sciences |
| spelling | doaj.art-0316747f277543a1b24c8ba84a9990162022-12-21T18:06:56ZengSpringerSN Applied Sciences2523-39632523-39712021-05-013611110.1007/s42452-021-04658-6Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordinationOgbonnaya Anicho0Philip B. Charlesworth1Gurvinder S. Baicher2Atulya K. Nagar3Liverpool Hope UniversityLiverpool Hope UniversityLiverpool Hope UniversityLiverpool Hope UniversityAbstract This work analyses the performance of Reinforcement Learning (RL) versus Swarm Intelligence (SI) for coordinating multiple unmanned High Altitude Platform Stations (HAPS) for communications area coverage. It builds upon previous work which looked at various elements of both algorithms. The main aim of this paper is to address the continuous state-space challenge within this work by using partitioning to manage the high dimensionality problem. This enabled comparing the performance of the classical cases of both RL and SI establishing a baseline for future comparisons of improved versions. From previous work, SI was observed to perform better across various key performance indicators. However, after tuning parameters and empirically choosing suitable partitioning ratio for the RL state space, it was observed that the SI algorithm still maintained superior coordination capability by achieving higher mean overall user coverage (about 20% better than the RL algorithm), in addition to faster convergence rates. Though the RL technique showed better average peak user coverage, the unpredictable coverage dip was a key weakness, making SI a more suitable algorithm within the context of this work.https://doi.org/10.1007/s42452-021-04658-6Swarm intelligenceReinforcement learningMulti-HAPSAutonomous coordination |
| spellingShingle | Ogbonnaya Anicho Philip B. Charlesworth Gurvinder S. Baicher Atulya K. Nagar Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordination SN Applied Sciences Swarm intelligence Reinforcement learning Multi-HAPS Autonomous coordination |
| title | Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordination |
| title_full | Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordination |
| title_fullStr | Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordination |
| title_full_unstemmed | Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordination |
| title_short | Reinforcement learning versus swarm intelligence for autonomous multi-HAPS coordination |
| title_sort | reinforcement learning versus swarm intelligence for autonomous multi haps coordination |
| topic | Swarm intelligence Reinforcement learning Multi-HAPS Autonomous coordination |
| url | https://doi.org/10.1007/s42452-021-04658-6 |
| work_keys_str_mv | AT ogbonnayaanicho reinforcementlearningversusswarmintelligenceforautonomousmultihapscoordination AT philipbcharlesworth reinforcementlearningversusswarmintelligenceforautonomousmultihapscoordination AT gurvindersbaicher reinforcementlearningversusswarmintelligenceforautonomousmultihapscoordination AT atulyaknagar reinforcementlearningversusswarmintelligenceforautonomousmultihapscoordination |