A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip Missions
The focus of unmanned aerial vehicle (UAV) path planning includes challenging tasks such as obstacle avoidance and efficient target reaching in complex environments. Building upon these fundamental challenges, an additional need exists for agents that can handle diverse missions like round-trip navi...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Drones |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-446X/9/1/26 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588659545604096 |
|---|---|
| author | Hyeonmin Kim Jongkwan Choi Hyungrok Do Gyeong Taek Lee |
| author_facet | Hyeonmin Kim Jongkwan Choi Hyungrok Do Gyeong Taek Lee |
| author_sort | Hyeonmin Kim |
| collection | DOAJ |
| description | The focus of unmanned aerial vehicle (UAV) path planning includes challenging tasks such as obstacle avoidance and efficient target reaching in complex environments. Building upon these fundamental challenges, an additional need exists for agents that can handle diverse missions like round-trip navigation without requiring retraining for each specific task. In our study, we present a path planning method using reinforcement learning (RL) for a fully controllable UAV agent. We combine goal-conditioned RL and curriculum learning to enable agents to progressively master increasingly complex missions, from single-target reaching to round-trip navigation. Our experimental results demonstrate that the trained agent successfully completed 95% of simple target-reaching tasks and 70% of complex round-trip missions. The agent maintained stable performance even with multiple subgoals, achieving over 75% success rate in three-subgoal missions, indicating strong potential for practical applications in UAV path planning. |
| format | Article |
| id | doaj-art-e1306eaaa1d24494bc771cfbbd97bb62 |
| institution | Kabale University |
| issn | 2504-446X |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Drones |
| spelling | doaj-art-e1306eaaa1d24494bc771cfbbd97bb622025-01-24T13:29:41ZengMDPI AGDrones2504-446X2024-12-01912610.3390/drones9010026A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip MissionsHyeonmin Kim0Jongkwan Choi1Hyungrok Do2Gyeong Taek Lee3Department of Industrial Engineering, Yonsei University, Seoul 03722, Republic of KoreaDepartment of Industrial Engineering, Yonsei University, Seoul 03722, Republic of KoreaDepartment of Population Health, NYU Grossman School of Medicine, New York, NY 10016, USACollege of Engineering, Gacheon University, Global Campus, Seongnam 13120, Republic of KoreaThe focus of unmanned aerial vehicle (UAV) path planning includes challenging tasks such as obstacle avoidance and efficient target reaching in complex environments. Building upon these fundamental challenges, an additional need exists for agents that can handle diverse missions like round-trip navigation without requiring retraining for each specific task. In our study, we present a path planning method using reinforcement learning (RL) for a fully controllable UAV agent. We combine goal-conditioned RL and curriculum learning to enable agents to progressively master increasingly complex missions, from single-target reaching to round-trip navigation. Our experimental results demonstrate that the trained agent successfully completed 95% of simple target-reaching tasks and 70% of complex round-trip missions. The agent maintained stable performance even with multiple subgoals, achieving over 75% success rate in three-subgoal missions, indicating strong potential for practical applications in UAV path planning.https://www.mdpi.com/2504-446X/9/1/26unmanned aerial vehiclefully controllable UAVpath planninggoal-conditioned RLcurriculum learning |
| spellingShingle | Hyeonmin Kim Jongkwan Choi Hyungrok Do Gyeong Taek Lee A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip Missions Drones unmanned aerial vehicle fully controllable UAV path planning goal-conditioned RL curriculum learning |
| title | A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip Missions |
| title_full | A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip Missions |
| title_fullStr | A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip Missions |
| title_full_unstemmed | A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip Missions |
| title_short | A Fully Controllable UAV Using Curriculum Learning and Goal-Conditioned Reinforcement Learning: From Straight Forward to Round Trip Missions |
| title_sort | fully controllable uav using curriculum learning and goal conditioned reinforcement learning from straight forward to round trip missions |
| topic | unmanned aerial vehicle fully controllable UAV path planning goal-conditioned RL curriculum learning |
| url | https://www.mdpi.com/2504-446X/9/1/26 |
| work_keys_str_mv | AT hyeonminkim afullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions AT jongkwanchoi afullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions AT hyungrokdo afullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions AT gyeongtaeklee afullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions AT hyeonminkim fullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions AT jongkwanchoi fullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions AT hyungrokdo fullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions AT gyeongtaeklee fullycontrollableuavusingcurriculumlearningandgoalconditionedreinforcementlearningfromstraightforwardtoroundtripmissions |